Modelling consumers' use of products

Patterns of consumers' use of products are of interest to manufacturers. This paper is concerned with modelling diary data on the use of one product, shampoo, recorded to the nearest hour by over 500 men during 1 week. A binary response multilevel model is developed, building on similar models for consumer purchase data. The model allows for a dependence on the number of days since last use. The results of fitting various versions of this model are discussed. A problem is that the number of days since last use is missing for all times up to the first use. An approximate EM approach is considered to deal with this problem.     

A three parameter model for birth projections

The approach underlying the model proposed in this article constitutes a considerable departure from the conventional methods used for the projection of birth series. Instead of following the customary procedure of directly projecting the age-specific fertility rates, they can be derived with this model from only three relatively simple fertility measures, namely, total fertility rate, mean age of fertility, and modal age of fertility. The reduction of the number of fertility parameters offers appreciable operational and analytical advantages. Among these, the most significant is the fact that statistical manipulation is confined to only three fertility measures, all of which are particularly appropriate for the in-depth analysis which is required to provide a rationale for assumptions of future fertility.     

Pearsonian Type I curve and its fertility projection potentials

New procedures are developed in this article for estimating parameters of the Pearsonian Type I curve which are particularly adaptive to factors influencing the pattern of age-specific fertility rates. It is shown that with this model the number of parameters required for the graduation and simulation of these rates can be reduced to only three—total fertility rate, mean and modal ages of fertility. The reduction in the number of fertility parameters offers considerable operational and analytical advantages, and makes the Pearsonian Type I curve particularly appropriate for the construction of a parametric model for fertility projections. In light of the results of empirical tests based on fertility data for Canada, the model’s potential for birth projections appears quite promising.     

Estimation Of The Birth Rate For The Congo Through Nonconventional Techniques

This paper attempts to produce a set of alternative estimates of birth rates for the Congo and for its main regions through nonconventional techniques. The main body of data used for this purpose is supplied by the demographic survey undertaken during 1955-57. Two types of information on fertility are produced by this survey: (1) births to women by age during the 12 months preceding the survey and (2) children-ever-born to women by age. The available data, while reliable in certain respects, are deficient in others. Thus, the reports on children born during the year preceding the survey are acceptably accurate for the children who have survived but are insufficient for the children who have died.The task was to make the best possible use of various reliable pieces of information directly or indirectly related to fertility, in order to produce some reasonably acceptable estimate of birth rate. One procedure consisted of correcting the reported birth rate for the omission of infant deaths; another procedure involved the adjustment of the reported birth rate with reference to the mean parity of women. In addition to these, an estimate of birth rate was derived by means of stable population technique from the proportion of children under 5 years of age.Three sets of estimates of birth rate, then, have been produced. They yielded a high agreement for the country as a whole, as well as for the provincial and district levels. These estimates were then compared with the proportion of childless women and the proportion of children-the latter having been derived from the records of "continuous registration of inhabitants," a source of information which is completely independent of the demographic survey of 1955-57. Again, a reasonable agreement has been found among all these estimates.Statistical evidence examined in this paper supports a birth rate of about 45 per 1,000 for the Congo. The analysis reveals a high regional variation in birth rate: from 20 per 1,000 (Bas Uele) to 60 per 1,000 (Kivu). This variation is consistent with the variation in sterility level as evidenced by the proportion of women never having had a live birth. The proportion of childlessness varies from 5 percent (Kivu) to about 40 percent (Bas Uele) for women past child-bearing age.     

Community Organization and the Oppressed

There is sexist bias in social work education, as teaching materials for the courses Human Growth and Development and Family Casework exemplify. Two leading authorities on adolescence, for example, accept a stereotypic view of female development. Family treatment writers reveal their ambivalence about male dominance within family structures, as they invoke such concepts as the “male head of the house,” and “sex role reversal,” while simultaniously encouraging independent development of each partner. In addition, the Women's Liberation Movement has created complex reactions among student practitioners.     

A three parameter model for birth projections

Abstract The approach underlying the model proposed in this article constitutes a considerable departure from the conventional methods used for the projection of birth series. Instead of following the customary procedure of directly projecting the age-specific fertility rates, they can be derived with this model from only three relatively simple fertility measures, namely, total fertility rate, mean age of fertility, and modal age of fertility. The reduction of the number of fertility parameters offers appreciable operational and analytical advantages. Among these, the most significant is the fact that statistical manipulation is confined to only three fertility measures, all of which are particularly appropriate for the in-depth analysis which is required to provide a rationale for assumptions of future fertility.     

Natality Estimates for the Canadian Indians by Stable Population Models, 1900–1969

Après avoir démontré que les données officielles sur les taux de naissance de la population indienne avant 1960 sont largement inadéquates, les auteurs utilisent deux méthodes indirectes pour obtenir des estimés fiables de ces taux. La première méthode se sert des donnees disponibles sur la distribution de la population par classe d'âge depuis 1900 tandis que la seconde utilise les données du recensement de 1961 sur l'ensemble des naissances selon l'âge de la mere. On analyse dans les deux cas, les données sous-jacentes et on déduit les estimés, à partir de modèles qui postulent une population stable. Les meilleurs estimés obtenus permettent de croire que le taux de natalité des Indiens est demeuré relativement stable dans le passé, au niveau de 50 naissances par 1,000 individus. On note cependant, un déclin constant de la fécondité depuis 1960. Ainsi le taux de natalité est passé de 46.0 pour 1000 en 1961 à 36.8 en 1969. Cette tendance peut être interpretee comme une evolution fondamentale dans la démographie de ce groupe ethnique. After having shown that official series of birth rates for the Indian population prior to 1960 suffer from considerable deficiencies, two indirect methods are applied to arrive at some trustworthy estimates of these rates. Use is made, in the one case, of the age distribution data available since 1900, and, in the other case, of 1961 census data on children ever-born to women by age. In both instances stable population models are used for analysing the underlying data and for deriving the estimates. According to the best obtained estimates, the birth-rate for the Indian population seems to have remained relatively stable in the past, at a level close to 50 per 1000 persons. A gradual decline in fertility has been taking place since about 1960. Thus the birth-rate has dropped from 46.0 in 1961 to 36.8 in 1969, the latest available figure. This trend may be regarded as a major development in the demography of this ethnic group which is, in many respects, unique in the Canadian context.     

Increase in natural fertility during the early stages of modernization: Canadian Indians case study

It has been hypothesized that populations may experience an increase in their natural fertility during the early stages of modernization as a result of the relaxation of various fertility-inhibiting practices and customs prevalent in traditional societies. This article offers evidence of such an increase in natural fertility among Canadian Indians. The main underlying cause is found to be in the massive, almost abrupt, shift from prolonged breastfeeding to bottle feeding which took place prior to the onset of large-scale birth control practices among Canadian Indians.