Correcting missing-data bias in historical demography

Studies on population history are often based on incomplete records of life histories. For instance, in studies using data obtained from family reconstitution, the date of death is right censored (by migration) and the censoring time is never observed. Several methods for the correction of mortality estimates are proposed in the literature, most of which first estimate the number of individuals at risk and then use standard techniques to estimate mortality. Other methods are based on statistical models. In this paper all methods are reviewed, and their merits are compared by applying them to simulated and to seventeenth-century data from the English parish of Reigate. An ad hoc method proposed by Ruggles performs reasonably well. Methods based on statistical models, provided they are sufficiently realistic, give comparable accuracy and allow the estimation of several other quantities of interest, such as the distribution of migration times.     

A frailty model for (interval) censored family survival data,applied to the age at onset of non-physical problems

Family survival data can be used to estimate the degree of genetic and environmental contributions to the age at onset of a disease or of a specific event in life. The data can be modeled with a correlated frailty model in which the frailty variable accounts for the degree of kinship within the family. The heritability (degree of heredity) of the age at a specific event in life (or the onset of a disease) is usually defined as the proportion of variance of the survival age that is associated with genetic effects. If the survival age is (interval) censored, heritability as usually defined cannot be estimated. Instead, it is defined as the proportion of variance of the frailty associated with genetic effects. In this paper we describe a correlated frailty model to estimate the heritability and the degree of environmental effects on the age at which individuals contact a social worker for the first time and to test whether there is a difference between the survival functions of this age for twins and non-twins.     

Perceived contra observed physical work load in Swedish dentists

In an earlier questionnaire study, dentists reported high experienced physical work load in their profession, but low to moderate complaints from their musculoskeletal system. The correlation between reported physical work load and pain from the musculoskeletal system was weak. This discrepancy could not be satisfactorily explained. Therefore, a second study was undertaken, in which the same 27 dentists who reported musculoskeletal problems were video recorded during one hour of clinical work, and the records were later analyzed using PEO (Portable Ergonomic Observation). PEO is a frequency analysis method which allows observation of work in real time using a portable computer or video recordings. PEO can be adjusted for registration of single or multiple work operations. Output data are presented as frequency, duration, and sequence of the various work operations. The aim of the present study was to investigate if there was a relation between observed work load recorded with PEO, and subjectively estimated work load and musculoskeletal complaints recorded with a questionnaire based on Visual Analogue Scales. Sitting and standing postures, and head, trunk and arm movements were analyzed. The PEO observations showed that dentists generally perform their clinical work in a sitting position, with the head bent forward almost half of the time. Only weak to moderate correlations (r =0.0-0.6) were found between observed physical work load and subjective estimations of experienced physical work load and musculoskeletal complaints. These findings support the results in our previous study, but they do not explain the large difference between the observed low work load and the subjectively experienced high work load. The study will be followed up by EMG measurements and free interviews, where both muscular load and psychosocial factors will be evaluated.     

Estimation of life expectancy in the Middle Ages

Summary. In the Middle Ages data on landholders were recorded. We use a selection of these data for estimating the life expectancy of adult males from the higher ranks of English society. The data consist of the age at which the landholder inherited land, a sequence of ages at which he was known to be alive and possibly the age at death. We explain that the date of death is left truncated and right censored at an unknown censoring time. We propose a model to fit (a selection of) the data, we give estimators of the unknown parameters in the model and we present estimates of adult life expectancy.     

Estimation of average mortality under censoring and truncation

In demography mortality is usually reported through averages over time intervals. If average mortality is estimated from censored or truncated data, then direct methods of estimation may create biases that depend on the censoring or truncation distribution. Such discretization errors may be avoided by estimating survival curves first in continuous time, and then discretizing the estimators. We illustrate the different methods on data of the form obtained from family reconstitution.     

Research Programming: Description of a Dutch Experience

In this article we discuss research programming as one of the means to bridge the gap between research and policy. Over the last decade research programming has come to be an integral part of science policy in The Netherlands. As an example of research programming put into practice we describe and evaluate the activities of the Programming Board for Research into the Social Services. We indicate the assets and liabilities of research programming.     

Maximum Likelihood Estimation of Life-Span Based on Censored and Passively Registered Historical Data

We consider the estimation of life length of people who were born in the seventeenth or eighteenth century in England. The data consist of a sequence of times of life events that is either ended by a time of death or is right-censored by an unobserved time of migration. We propose a semi parametric model for the data and use a maximum likelihood method to estimate the unknown parameters in this model. We prove the consistency of the maximum likelihood estimators and describe an algorithm to obtain the estimates numerically. We have applied the algorithm to data and the estimates found are presented.