Appendix I 



Methodology 



The general equation followed in estimating migration can be sym- 

 bolized as follows: Initial population, plus births, minus deaths, plus or 

 minus migration, equals final population. 



Migration by Age Groups 



In order to estimate migration by age group it was necessary to obtain 

 initial and final population by specified age groups and to estimate the 

 number of deaths occurring to this age group over the period of time in 

 question. The problem was one of tabulating resident deaths by single 

 years of age for each year of the period under study. 



It was possible to obtain resident deaths by single years of age in 

 New Hampshire from 1940 to 1950 from the State Department of Health, 

 Division of Vital Statistics. The recorded deaths were mechanically sorted 

 for each year according to the following criteria: 



1. Place of Residence: N. H. residents who died in the state or 



out of state were included. Deaths of non-residents occur- 

 ring in the state were excluded. 



2. Age of Deceased: Excluding a few whose age was not stated. 



3. County of Residence of the Deceased: Excluding a few whose 



residence was not stated. 



Since the population is enumerated as of April 1, it was necessary to 

 take the 1940 deaths from April 1 to December 31, and the 1950 deaths 

 from January 1 to March 31, to coincide with the census taking. 



It must be realized that the method of pairing deaths within an age 

 group to population in that age group is not wholly accurate. An example 

 will clarify this point. If an individual was aged 34 years, 11 months in 

 April, 1940, he would be enumerated in the 30-34 age group. However, if 

 he should die in June of the same year, his age at death would be 35. 

 This would cancel out if we assume that a 29-year, 11-month individual 

 may also die at the age of 30. In the higher age groups, however, with in- 

 creasingly higher-age specific death rates, the cancellation may not be wholly 

 complete. 



It is also evident that deaths to residents between April, 1940, and 

 April, 1950, will include deaths to in-migrants who are now residents, but 

 who were not enumerated in 1940. This intrinsic factor would tend to slight- 

 ly exaggerate net migration. 



Since the population in 1940 of any age group is known from the 

 U. S. Census, it is possible, by subtracting the deaths in the intervening 

 period, to estimate the expected population in 1950. The difference between 

 the expected population in 1950 and the actual census figure is an approxi- 

 mation of the net migration which has occurred over the period to the 

 age group. 



17 



