Volume 53 - Number 1 - Spring 1987

N. Methods of Research and Analysis Including Models

Studies concerned with demographic methods and with methods from other disciplines that have been applied to demographic data as a whole. Includes mathematical demography and studies on methods of estimation and indirect estimation. Methodological studies and models concerned with one demographic variable, such as migration, are coded under the category concerned with that topic and cross-referenced to this heading. Studies on models used to investigate relationships between demographic variables and for the analysis of empirical data are also coded under this heading.

53:10780 Alonso, William. Gravity models. Center for Population Studies Discussion Paper, No. 86-9, Dec 1986. 6 pp. Harvard University, Center for Population Studies: Cambridge, Massachusetts. In Eng.
The author reviews the gravity model developed by Stewart in the 1940s to describe observed spatial regularities in populations. Extensions of the model and subsequent applied works are traced, and the significance of these types of models is assessed.
Location: Princeton University Library (SPR).

53:10781 Alonso, William. Systemic and log-linear models: from here to there, then to now, and this to that. Center for Population Studies Discussion Paper, No. 86-10, Nov 1986. 32 pp. Harvard University, Center for Population Studies: Cambridge, Massachusetts. In Eng.
The author seeks to "make explicit the link between the systemic formulation of functional models [for analyzing transition matrices] and an important family of statistical models, namely those employing a loglinear approach to the analysis of variance." He proceeds by summarizing the two types of models, integrating them, and suggesting some applications.
Location: Princeton University Library (SPR).

53:10782 Beaumont, Paul M. ECESIS: an interregional economic-demographic model of the United States. Pub. Order No. DA8505038. 1984. 504 pp. University Microfilms International: Ann Arbor, Michigan. In Eng.
"The ECESIS model examines the linkages between interregional economic and demographic models and determines the importance of those linkages in simulation and forecasting experiments. Structural econometric models are built for all fifty states plus the District of Columbia....This interregional economic system is linked to an interregional cohort-component demographic accounts model. The accounts model provides age and sex detail for all fifty-one regions and estimates all state-to-state migration flows." It is found that "the economic-demographic linkages do not change the qualitative properties of the model but do have a significant quantitative impact. It is also found that the system is quite sensitive to differing migration submodels but not too sensitive to the demographic accounting system itself. The model is used to examine the impacts of regional investment tax credits to stimulate growth in depressed areas and changes in U.S. immigration policy."
This work was prepared as a doctoral dissertation at the University of Pennsylvania.
Source: Dissertation Abstracts International, A: Humanities and Social Sciences 46(1).

53:10783 Clairin, Remy; Conde, Julien. Handbook for estimating demographic statistics from incomplete data in developing countries. Development Centre Papers, Pub. Order No. 38.353. Sep 1986. 266 pp. Organisation for Economic Co-operation and Development [OECD], Development Centre: Paris, France. In Eng.
This manual is designed to provide the demographer working with incomplete data with information concerning the basis for the theoretical methods and techniques at his or her disposal for demographic analysis, the calculations needed to apply such methods and techniques, and the mortality models available. The emphasis is on the practical application of such methods and techniques to actual data. The publication is in three parts: the first part is concerned with the mathematical relations involved, the second part with applications to observed data, and the third part with the main model life tables available. The geographic focus is on developing countries
This document is obtainable on request from the Information-Publications Service, OECD Development Centre, 94, rue Chardon-Lagache, 75016 Paris, France.
Location: Princeton University Library (SPR).

53:10784 Gray, Alan. The missing ages: adjusting for digit preference. Asian and Pacific Population Forum, Vol. 1, No. 2, Feb 1987. 11-22 pp. Honolulu, Hawaii. In Eng.
The author sets forth a method for adjusting age distributions to eliminate the effects of digit preference and seeks to overcome a bias present in an approach developed by Zelnik in 1961. Census age pyramids are derived for Bangladesh, Libya, and the Australian aborigines for the years 1974, 1973, and 1981, respectively.
Location: Princeton University Library (SPR).

53:10785 Inaba, Hisashi. On the discrete model of multiregional demographic growth. Jinko Mondai Kenkyu/Journal of Population Problems, No. 179, Jul 1986. 1-15 pp. Tokyo, Japan. In Jpn. with sum. in Eng.
A discrete, one-sex model of multi-regional demographic growth is developed. The author considers the conditions under which this demographic process is strongly ergodic.
Location: Princeton University Library (SPR).

53:10786 Isupov, Arkady. The United Nations manuals for population analysis. Population Bulletin of the United Nations, No. 19-20, 1987. 90-6 pp. New York, New York. In Eng.
Eight manuals on population analysis, evaluation, and forecasting, published by the United Nations between 1952 and 1983, are reviewed. A brief overview of the subject matter of each manual is provided. "The primary purpose of the manuals was to help countries to obtain population information and to assess its quality. It was also intended that the manuals might be used as teaching materials in the training of population specialists." The author suggests that given developments in population statistics and techniques of data calculation, revisions and updating are needed.
Location: Princeton University Library (SPR).

53:10787 Keilman, Nico; Gill, Richard. On the estimation of multidimensional demographic models with population registration data. Working Papers of the NIDI, No. 68, Feb 1986. vii, 34 pp. Netherlands Interuniversity Demographic Institute [NIDI]: Voorburg, Netherlands. In Eng.
"In this paper the estimation of multidimensional demographic models and their underlying stochastic processes is investigated in situations where population registration data are available. With this kind of aggregate data, estimation by traditional methods is not possible. We look at two versions of the multidimensional model: the constant intensities model and the linear integration model. Some logical inconsistencies in the derivation of the latter are discussed. In particular, we argue that the linear integration model is not compatible with a Markov process. A new estimation algorithm for the estimation of the constant intensities model with population registration data is proposed. Some preliminary results on the mathematical and statistical properties of this method are given. The algorithm is applied to Dutch nuptiality data."
Location: Princeton University Library (SPR).

53:10788 Keyfitz, Nathan. Heterogeneity and selection in population analysis. Statistics Canada Research Paper, No. 10, Sep 1984. 30 pp. Statistics Canada, Research and Analysis Division: Ottawa, Canada. In Eng.
"This paper addresses two particular aspects of heterogeneity which are of considerable direct relevance to social science and statistical analysis. One is the effect of heterogeneity on the error structure of data. The other is the relation between heterogeneity and the representativeness of certain kinds of aggregate measures." After a discussion of the impact of heterogeneity on the assessment of error in data, the author considers heterogeneity and selection over the life course, with attention to mortality and notions of frailty over time.
Location: Princeton University Library (SPR).

53:10789 Lam, David. Lorenz curves and inequality comparisons under changing population composition. Population Studies Center Research Report, No. 86-92, Apr 1986. 11 pp. University of Michigan, Population Studies Center: Ann Arbor, Michigan. In Eng.
"This paper analyzes the properties of Lorenz curves produced from combinations of income distributions. The motivation for the analysis is that many demographic and economic changes can be modeled as the introduction of new entrants into some base population. The paper shows that when the sub-populations have equal means, a number of simple conditions govern the position of the Lorenz curve for the combined population relative to the Lorenz curves for the two sub-populations. These conditions become less regular when the means of the sub-populations differ, and suggest that a combined distribution will almost never Lorenz dominate an original distribution. The results point out an important source of ambiguity in intertemporal inequality comparisons."
Location: Princeton University Library (SPR).

53:10790 Mukerji, S. Some methodological issues in demography. In: Dynamics of population and family welfare, 1985, edited by K. Srinivasan and S. Mukerji. Dec 1985. 41-71 pp. Himalaya Publishing House: Bombay, India. In Eng.
The author evaluates and discusses the limitations of four methods frequently used in demographic analysis. "The first one deals with the application of the regression method used for the study of selected predictor variables on fertility, or the couple protection rate, or any other demographic parameter to be investigated....The second statistical technique, which has been assessed and commented upon by the author, is the well-known 'recall lapse theory', originally proposed by R. K. Som and others, and later used very widely to correct the data of...surveys to adjust for recall lapse on children reported born by the respondents for the last period of the reproductive life."
The author then discusses "the limitations in the use of the concept of [the] infant mortality rate for the purposes of drawing inferences on comparative health conditions of the population, or for studying trends in the health conditions over time." Finally, the author critically evaluates the use of place of birth data to study internal migration. Data for India are used to illustrate the arguments.
Location: Princeton University Library (SPR).

53:10791 Naidenova, Penka. The population problem in global modeling. [Problemat za naselenieto pri globalnoto modelirane.] Naselenie, Vol. 4, No. 2, 1986. 34-52 pp. Sofia, Bulgaria. In Bul. with sum. in Eng; Rus.
Developments during the past 15 years in population modeling are critically reviewed. The author notes that while population variables were treated as endogenous in earlier models developed by the Club of Rome, later models have treated such variables as exogenous. The need to link demographic factors to structural changes and economic growth, in accordance with Marxist-Leninist population theory, is noted.
Location: Princeton University Library (SPR).

53:10792 Preston, Samuel H. The relation between actual and intrinsic growth rates. Population Studies, Vol. 40, No. 3, Nov 1986. 343-51 pp. London, England. In Eng.
"This paper demonstrates that, to a close approximation, the intrinsic growth rate of a population is equal to the mean of age-specific growth rates below age T, the mean length of a generation. This mean is normally close to the growth rate of the entire population block below age T. Therefore, when a disparity exists between the intrinsic growth rate and the actual growth rate of a population (whether or not net migration is included in both rates), it must be attributable to an unusual growth rate of the population block above age T. One implication is that the 'momentum of population growth' is entirely confined to the age span above T, approximately age 28 in developing countries today."
Location: Princeton University Library (SPR).

53:10793 Shah, M. Rafiq. The use of a piecewise constant proportional hazards model in issues related to standardisation: a review and some results. IPD Working Paper, No. 1987-2, 1987. 36 pp. Vrije Universiteit Brussel, Interuniversity Programme in Demography: Brussels, Belgium. In Eng.
The author examines the piecewise constant proportional hazards model and its relation to standardization. "First, the definition of a piecewise constant proportional hazards model will be followed by the derivation of a piecewise log-likelihood function (log L). Then it will be shown that (log L) is equivalent to the one obtained from observations having a Poisson error structure in a contingency table perspective. Later on, a log-linear model will be fitted on the data used by Hoem (1979) by using GLIM (Generalized Linear Interactive Modelling...)for the purpose of illustration and new interpretation of the parameters of a classical multiplicative model. Finally, the use of the proposed model will be commented in the light of various issues of standardisation; in particular to examine the claim that the proposed model solves the issues of both direct and indirect standardisation."
Location: Princeton University Library (SPR).

53:10794 Silcocks, P. B. S.; Murphy, M. Relative risk estimation from vital statistical data: validation, a pitfall and an alternative method. Journal of Epidemiology and Community Health, Vol. 41, No. 1, Mar 1987. 59-62 pp. London, England. In Eng.
"A previously described method of obtaining an estimate of relative risk from routinely available data was applied to data on cigarette consumption and lung cancer mortality to test its validity. Some shortcomings of the method were noted and an alternative approach using weighted logistic regression gave results closer to those predicted on the basis of other studies, without the disadvantages of the original technique." The data concern England and Wales, are from official sources, and pertain to the period 1969-1973
Author's address: Department of Clinical Epidemiology and Social Medicine, St George's Hospital Medical School, Cranmer Terrace, London SW17 0RE, England.
For the method referred to, by Valerie Beral et al., published in 1979, see 46:3848.
Location: Princeton University Library (SPR).

53:10795 Soeradji, Budi. A sampling frame for population research. [Kerangka sampling induk untuk penelitian kependudukan.] Majalah Demografi Indonesia/Indonesian Journal of Demography, Vol. 13, No. 25, Jun 1986. vii-viii, 21-30 pp. Jakarta, Indonesia. In Ind. with sum. in Eng.
Questions concerning the selection of an appropriate sampling frame for the study of population questions in Indonesia are considered. The author suggests that the sampling frame from the census is inadequate for reasons of both size and data availability. It is recommended that using the results of a large demographic survey, such as the SUPAS Intercensal Survey of 1985, would yield better results.
Location: Princeton University Library (SPR).

53:10796 Vanderhoeft, C. Hidden heterogeneity in composite link models: further developments. IPD Working Paper, No. 1987-1, 1987. 32 pp. Vrije Universiteit Brussel, Interuniversity Programme in Demography: Brussels, Belgium. In Eng.
The author seeks to provide an overview of composite link models and the problem of taking into account unobserved heterogeneity. "Attention is focused mainly on models for binomial and Poisson data. In Section 2 we discuss an extension of Thompson and Baker's (1981) composite link model. In Section 3 we introduce in this model multidimensional hidden heterogeneity (for binomial and Poisson data). In Section 4 and Section 5 we discuss approximate models which allow for estimation of the parameters of the composite link model with hidden heterogeneity. Section 6 applies the methods in a reanalysis of binomial data used by Williams (1982) and Poisson data used by Breslow (1984). Finally, Section 7 introduces a simulation study."
Location: Princeton University Library (SPR).

Copyright © 1987-1996, Office of Population Research, Princeton University.