DEMOGRAPHIC TREATMENT OF KINSHIP

• Main Author: K. Sembiring, Robert
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/788
• Institution: Institut Teknologi Bandung

<b>ABSTRACK:</br></b> <br /> <p align=\"justify\">An experiment has been done on the relation of birth and death rates on the one hand to the expected number of living kin and the probability of a direct progenitor still being alive on the other. All through the female line (one-sex model). This experiment is an application of the theory worked out by L. A. Goodman, N. Keyfitz, T. W. Pullum. <br /> By assuming a stable population model, several hypothetical populations were constructed. In addition, four \"actual\" populations, similarly assumed to be stable, were also included in the study. It was found that fertility makes more difference than does mortality to the expected number of living direct descendants. The same holds true with the expected number of living sisters, nieces, aunts, and cousins. However, mortality makes more difference than does fertility to the probability of a direct progenitor being still alive. <br /> A multiplicative model of the form <br /> Y = a0 .(GRR) al .(ee o) a2 a3, <br /> where GRR is the gross reproduction rate, 8o is the expectation of life at zero, and p is the mean age of childbearing in the stable population, was found to fit very well to the data on each of the various kinship relationships. <p align=\"justify\">This model then was used to predict the various kinship relationships of the United States, 1967 from those of the Unites States, 1959-60. The results are satisfactory for expected number of living direct descendants, sisters, nieces, aunts, and cousins but unsatisfactory for the probability of a direct progenitor being still alive. Several perturbations on the input variables were also considered. The results, whenever possible, were compared with those obtained using the multiplicative model. It was found that for the expected number of living kin the agreement between the results obtained from the perturbations and those obtained using the multiplicative model is substantial. However, for the probability that a direct progenitor is still alive the agreement is poor. Backward interpolation method was employed to infer 80 and NRR. The method provides a satisfactory solution to some of the integral equations of the various kinship relationships.