Bad Science is a book by Ben Goldacre , Fourth Estate, London, 2008 and stems from a column he writes in the Guardian newspaper called Bad Science and you can read his blog at http://www.badscience.net. Mr. Goldacre writes about your health and what you should or should not believe. A great deal of good health has to do with eating the basic food groups, do regular exercise, and display the traits of a good person on a regular basis. Since some people get sick and others seemingly doing the same thing do not, nobody really knows what is the best thing. That does mean various industries have not come forward and recommended things that you could do to improve the likelihood of not getting sick or doing preventive actions. How they do that, is what Mr. Goldacre writes about – and part of the task is to be cynical about the answer given by companies with the desire to make money from the process. The process the health industry and many other industries use to help answer the question are remarkably the same:

Use of Experts or Authority – anyone who spends a great amount of time on something will become an expert. If you wish to be paid for spending a great amount of time on the issue, you will need some initials behind your name. Every professions spends time lobbying the government to ensure the initials the professional body gives remains the top choice of the government departments which most interact with it. For example if you wish to audit tax returns your designation has to be CA or Charter Accountant. Other designated accountants can do the same thing but will not be recognized by the government or recognized by customers. In the case of health – we as a general public expect something similar to Doctor, because the Doctors have lobbied for the past couple of hundred years, if it health related, you need a Doctor. This does not mean all Doctors are great, and we hope your doctor is one of the great ones, but it does mean for health advice – doctor is the bedrock of advice.

Show Statistics of Improvement – if we do something we hope that the new is better than the existing method. Every year, thousands of new graduates come into the marketplace with great potential and something new and better will come from them. They will need to show the problems they are trying to solve can be solved with something new. The classical tricks to play in the statistical analysis in order positive results shows up include:

Ignore the Protocol Entirely – always assume that any correlation proves causation. Place all the data and if you have measured enough some things are bound to be positive by sheer luck.

Play with the Baseline – sometimes a trial starts and it is doing better than the trial on the placebo or sugar pills group. If so leave alone, if the reverse adjust the baseline in the analysis to show the positive results.

Ignore Dropouts – in all trials of people, some people will not come every week or will drop out of the surveys. Do not report them because they have likely faced some side affects to the drugs.

Clean up the data – all results are put on a graph. If the results from the ones outside of the major concentration of points or outliers help your data leave them in, if not drop them.

Torture the Data – if the results are not showing what you want figure out what good things it shows and concentrate on those results.

Linking to dividend paying stocks, the analysis done in the medical profession for new drugs is similar to the investment industry new products each year. In terms of investments when markets go down, there is an opportunity that it will go back to the normal trading range, (hopefully you did not own the stocks when it went down), new products will emerge to invest in the new opportunity. The reality is over the long term the best stocks – those that consistently make a profit which allows them to pay a dividend is the only ones you have to concentrate on. If the market goes down, you will have dividend payments, if the market goes up, the market pays higher multiples for profitable stocks.

There are more questions than answers, till the next time – to raising questions