The April edition of the Langport Leveller includes data from the Office of National Statistics on life expectancy. Tables showing the electoral wards with the highest ten and the lowest ten life expectancies in the county reveal an unexpected disparity. In the Mendip village of Beckington, where people live some of the longest lives in the country, the life expectancy is 95.6 years. Twenty-five miles away, in the Saint Benedict’s area of Glastonbury, people might expect to live only 76.1 years. The Saint Benedict’s figure is the 24th lowest in the country, it is an expectancy that is comparable with some of the poorest urban areas in the country.

If one were to accept the statistics as they appear, one might assume there existed some significant difference between the two community. Would there be environmental conditions, or social problems, or high levels of deprivation in Glastonbury, or other complex disorders that would explain the reason why people in the town live nineteen years less than the people of Beckington? Or is it the other way, is there something in the water in the Mendip Hills that allows people to expect to enjoy a life lasting almost a century, or something in the air that keeps them healthy, or fruit and vegetables in the gardens which provide some special elixir?

If not a case of lies, damned lies and statistics, the gap in life expectancy between Glastonbury and Beckington is a case of needing to beware of outliers. The figure for the village is that for a small community, in a small sample, outlying figures can distort averages. A handful of long-lived families can create the impression of a village where to be a centenarian is a reasonable expectation. The figure for Saint Benedict’s ward in Glastonbury is also presumably a statistic that owes much to outlying cases that create a picture of the place very different from that known to local people and encountered by countless people.

Contemplating a move to Beckington to ensure value from my pension fund, I remembered my grandmother died eleven years ago at the age of ninety-four, that both my parents are still alive, and that I should have a reasonable chance of attaining advanced years. I also remembered the statistical concept of regression toward the mean, that extreme variables tend to move back toward th average, the next time they are measured. The next statistical table showing the respective life expectancies of Beckington and Glastonbury Saint Benedict’s will undoubtedly show that both have shifted close to the average figure for the county. There are no secrets, just a statistical snapshot that should only be read in the context of all the other information available.