Whenever I’m faced with some statistical evidence offered in opposition to an issue in my case, I think about the conclusions from a survey done in 1936. A well-known magazine of that era, The Literary Digest, had funded a national poll in order to predict the outcome of the presidential race between Roosevelt and Landon. The names of the people who were to be polled were taken from telephone listings and automobile registration lists. A very large number of people, some 2 million, were polled during the survey. The results clearly predicted a very wide victory margin for Landon over Roosevelt. When the votes were counted however just the opposite was the case and Roosevelt won.

What happened? It turns out that in 1936 only a small number of the population of the United States owned either telephones or cars and most were in urban areas that favored Landon. Therefore, while a very large number of people were surveyed the great majority were the wealthy and those who live in cities where Landon had strong support and excluded the rest of the population that lacked cars or telephones. It was a biased survey with a false conclusion passed off as scientifically accurate. With corrupted medical research and biased statistical analysis we are more often than not faced with invalid statistical arguments. Which means that we should not accept research conclusions and survey findings simply because they are presented as if they were scientifically valid. Look behind the conclusion to the method employed in arriving at the conclusion and always check both the size of the sample used as well as the method employed.

My friend Eric Oliver is an excellent jury consultant whose company is known as Metasystems, which is a reflection of Eric’s Neural Linguistic Programming background. His newsletter is available at Eric-Oliver.com  is one I enjoy reading. I think Eric would forgive me for some minor plagiarism of miscellaneous ideas which I picked up from the last issue, Fall 2012.

In an article about anchoring, Daniel Kahnerman him wrote about a study done with a rigged the wheel of fortune marked from 0 to 100 but built so that it would stop only at 10 or 65.  Students at the university where it was conducted were asked to spin the wheel and then write down the number on which the wheel stop , which was either 10 or 65 . After writing it down, they were then asked some questions to estimate the percentage of African nations which they knew the students would not know and would have to estimate.  What the researchers  found was the higher the number on the wheel the student had written down, the higher the estimate and vice versa. Research has established that the reason was because of the number having an anchoring effect and also because the number acted as a primer as well as an anchor in influencing the guess at the answers to the questions.  In other words the suggestion (by the wheel number) acted as a primer to influence the opinion given as well as being an anchor for the estimate..

What is the practical significance of this? For one thing opening numbers have an influence in litigation. The article uses the example of the purchase price of a house. Buyers are influenced by the asking price. The listing price influences our  attitude about the value of the house. Take jury selection, for example. In jury selection if you tell the jury you have read about a verdict of $15 million dollars that seemed excessive to you and then ask what amounts the jury thinks are excessive, their responses will be influenced by the number you offered first. It also has an anchoring effect on their thinking about case values. Take negotiation. In negotiation the article suggests that if you think the other side has made an outrageous proposal you should not come back with an equally outrageous counteroffer, creating a gap that would be difficult to bridge or allow your self to be in some way influenced by it. Instead, the writer says,  you should refuse to continue the negotiation with the number in that range until a more reasonable one is offered.

in another article, Sasha Issenberg writes about political issues. He makes the same point that linguist George Lakoff has made in his writings, namely that people act out and vote according to moral values which Lakoff says trump even one’s personal interests. Lakhoff has argued that a trial is really a battle of right and wrong from the jurors personal belief system. Eisenberg says that when we talk about our beliefs regarding welfare or abortion we are simply using borrowed language to express one of six moral values we hold namely: harm, fairness, loyalty, authority, liberty, and sanctity.  People become involved on a political level by the moral matrix involved .  When we are dealing with jurors  our case issues are being evaluated through evidence and testimony by moral filters. Knowing a juror’s deeply held values is more important that knowing demographic information about the juror.

Those a some rather basic thoughts you probably already knew, but which are helpful to remind ourselves about as well.