This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 1 John Graunts Life Table In 1661 A.D., some 3,064 years after the first census of Moses, an obscure haberdasher, late a captain in the loyalist army of Charles II, published an analysis on data originally collected by Thomas Cromwell, 127 years earlier, dealing with age at the time of death in London. The data had been collected at the request of the merchants of London whowere carrying out the most basic kind of marketing research: i.e., whether potential customers (i.e., live people) were on the increase or decrease. Interestingly, enough, the question originally arose because of the fact that the Plague (perhaps the Bubonic Plague, perhaps a collection of various contagious diseases) had been endemic in England for many years. At times, there would be an increase of the incidence of the disease,at other times a decrease. It was a matter of sufficient importance to be attended to by Chancellor Thomas Cromwell (also Master of the Rolls). Without any central data bank, a merchant might put a shop in an area where the decline in population had eliminated any potential opportunity, due to market saturation. Cromwells data base consisted in records of births and deaths from the Church of Englandto be carried out and centrally lo- cated by the clergy. Before Graunt, all analyses of the data had suffered the usual cant see the forest for the trees diffi- culty. The records were not kept in EXCEL. Each parish priest (and there were hundreds) had his own way of recording births, deaths, marriages, etc. The data base was included in ledgers without any good sense of a common taxonomy. Graunt solved this problem, and started modern statistics by creating the following table: John Graunts Life Table Thompson 2 Table 1. Graunts Life Table. Age Interval Prop. Deaths in Interval Prop. Surviving til start of Interval 0-6 0.36 1.00 7-16 0.24 0.64 17-26 0.15 0.40 27-36 0.09 0.25 37-46 0.06 0.16 47-56 0.04 0.10 57-66 0.03 0.06 67-76 0.02 0.03 77-86 0.01 0.01 Graunt did a little too much smoothing, for we only know the death proportion in each of the intervals. But Graunt could stilldeath proportion in each of the intervals....
View Full Document
This note was uploaded on 02/04/2011 for the course PB HLTH 140 taught by Professor Tarter during the Fall '10 term at University of California, Berkeley.
- Fall '10