Showing posts with label statistics. Show all posts
Showing posts with label statistics. Show all posts

Science and Sensibility

Times change. People change. It seems that nowadays facts change too.

To start with a very simple example, I have heard changing medical advice on whether we should drink the glass of water during, right after or half an hour after meals. Daily walk is good for the health – but do you walk slowly or briskly, before breakfast or after dinner? Some do say coffee is good for the heart, still others say wine has its benefits. There are very recent studies to suggest harmful effects of cow milk!

Look at the never ending debate on whether Daylight Savings Time saves electricity. Now you would think that electricity usage was both measurable and also accurately model-able but still there is no definitive outcome of the different studies.[timeanddate.com]

I am sure life expectancy has increased over the decades. But so has the number of fatal diseases and the percentage of people suffering from them. One might argue that the diseases were not detected peviously. In my observation though the grandparent generation lived much longer and healthier than their following generations. Think about how many things today cause cancer. The more responsible texts throw in a subdued politically correct phrase like 'have been found to increase the chance'. Basically doctors still don't actually know what causes cancer and having to lay the blame somewhere, they suggest lifestyle guidelines one or the other of which you are not likely to follow. The speculative attributing of reasons of cancer started with a face-off on the then-newer products like cellphones[WHO] and microwave ovens. The list kept growing to include smoking, alcohol, pollution, sunlight, red meat, fat dairy products, age, genetic make, and so on and on [ACS]. Also fish and vegetables are not safe because of the preservatives, fertilizers or illicit chemical colorings applied. It is almost as if - if you live, you get cancer!!

Smoking is, to borrow from Fletcher Knebel, the leading cause of statistics. It causes everything from TB to heart attacks to cancer:
- One in two lifetime smokers will die from their habit [Better Health Channel].
- Tobacco is responsible for 18.1% of total deaths in 2000 [American Medical Association].
- More deaths are caused each year by tobacco use than by all deaths from HIV, illegal drug use, alcohol use, motor vehicle injuries, suicides and murders combined [FDA].

The broad availability of such sporadic 'statistics' makes me skeptic. Will the non-smoker not die?! What's a 'life-time' 'smoker'? In the comprehensive (nationwide, sub-parts adding up to 100%) data on causes of death provided by CDC's National Vital Statistics Report (Table 9, Page 27), smoking is not listed. So obviously the claimed effects of smoking is indirect – it influences (not causes, as reason is still unidentified) one or more of the causes of the diseases. Now then, how do these researchers figure out whether the cancer or heart disease in a certain patient is influenced by tobacco and not by, say, Genetically Modified Foods?

Given that the history of smoking dates back to as early as 5000–3000 BC, added to the modern curses of stress and pollution, the human race would be extinct by now! And if cigarettes are so bad, why is it not banned? Cigarette manufacturers are not even required to give out information to the public about the ingredients and additives used in cigarettes. Read more about the anti-smoking propaganda in The Smoking Issue-An essay by Joe Jackson. Nothing in this world is simple anymore, vested interests reign everywhere.

Let's move beyond to the topic of the times. Climate Change. Mind you at the onset, it is all prediction. And how many predictions can scientists claim to have made correctly in the past? We cannot even make weather forecasts a few day's ahead. So how much of a base does a long term future analysis of not a particular region but the entire globe have?

What is causing this greenhouse effect - the use of cars, industries and inefficient fuel utilization therein. So after all the forward progress with manufacturing engines and making highways, we are not to use them - irony indeed. New causes of greenhouse effect have come out and would you believe that they include water vapor, eating meat and having pets:
- Livestock and their byproducts account for at least 51% of annual worldwide GHG (greenhouse gas) emissions [World Watch Institute].
- The domestic cat population of the ten largest feline-friendly nations takes 154,400 square mile of agricultural land to support them [New Scientist].
What do you do with these figures? Turn to herbivores? Are cows, cats and humans not part of nature? Rather, what is not a part of nature? We are not the creator and so we have not 'created' anything. Why should nature not be supportive of human civilization? Is saving endangered species not interfering with nature? Should we go back to living in the jungles but even then every breath we take would interfere with the atmosphere.

I am not saying smoking or pollution is good. I just do not need dramatic numbers to use common sense along with some simple basic principles that we are taught since childhood, like Waste Not, Want Not.

So why this mad desperate rush for fast new 'findings' for which a lot of people are putting forward these embellished statistics? It could be aided by the lack of breakthrough scientific inventions/discoveries in the 21st century. Perhaps physics has seeing the evolution of quantum particle theories and biologists are grasping the first level understanding of DNA, but what else has happened of late? And even before that most discoveries/inventions have been accidents. Sure our lifestyle is advancing – we are reaping the benefits of commercial application of the past inventions. The internet and radio communication was there much before Facebook and cell phone. Stephen Hawking, in A Brief History of Time, guesses that the existing knowledge base has grown to become too large for a single mind to learn/understand and then to go beyond to come up with something new.

Perhaps that is the case. Or perhaps as a easier method, a lot is increasingly 'based' on statistics [Odds Are, It's Wrong]. The term has a scientific touch to it. Also it is easier in the sense that you observe rather than predict. You choose your sample size and control parameters and derive at the conclusions that have no way of being proved or disproved. Still you get a theory, for mind the definition: A theory, in the scientific sense of the word, is an analytic structure designed to explain a set of empirical observations.

Saying NO to DST

http://www.thedailystar.net/newDesign/news-details.php?nid=90815

The first thing that struck me about the news was that we are set to start DST without deciding how long it is to continue. The overall ease with which this decision was taken amidst so many other pressing national concerns shows that not much consideration was given to it. This bothered me as I always considered Daylight Saving to cause much unnecessary confusion and hassle.