Demographics and Statistics

views updated

Demographics and Statistics

Julius Richmond, the former Surgeon General of the United States, is purported to have said, "Statistics are people with their tears wiped dry" (Cohen 2000, p. 1367). While it is true that statistics, and quantitative data more generally, have a "dry face" to them, they have important uses in research and public policy. Statistical and demographic data are not meant to provide understanding on the felt circumstances of individuals. By their very nature these data deal with social aggregates.

Although people think that quantitative data give an objective portrayal of a phenomenon (the facts), this is not correct. What researchers choose to be measured and the methods they employ reflect the biases and values of those who collect data. Mortality data are almost always collected by official or government agencies; thus to greater or lesser degree they reflect their perspectives. However, some measures of mortality, in particular causes of death, have been "internationalized" by such bodies as the World Health Organization and therefore reflect a consensus, albeit a Western-based one. In addition, some developing countries do not have the resources to acquire very much data on demographic events such as deaths; if they did have the available resources, it is not known what kind of information they might collect.

What is chosen to be measured and how it is measured is only part of the bias in quantitative data, however. How data are interpreted is also subject to bias and value judgments, clearly seen, for example, in the debate about the factors leading to maternal deaths and how to reduce maternal mortality.

Apart from biases, users of quantitative data on deaths need to be aware of a number of limitations. A large limitation, globally, is simply lack of information. Many statistics are estimates only. Another limitation concerns lack of knowledge regarding how statistics are calculated, which can lead to misinterpretations. A good example of this is with statistics on life expectancy which, although hypothetical, are not always interpreted as such.

Statistical data provide important information that is useful for a number of purposes, despite their limitations, problems with bias, and an inability to convey individual experiential phenomena. Scientists and researchers need to know how many people are dying and at what ages, of what gender, and for what reasons, in order to know how to target resources to reduce those deaths. Unlike the case with other demographic topics such as fertility and migration, there is worldwide consensus that reducing deaths is a worthwhile goal; thus statistical data on mortality can be corroboratively used in attempts to reach that goal. Data provide the raw materials needed for plans to be made (and implemented) aimed at enhancing the wellbeing of persons and tackling social inequalities in the risk of death.

See also: Causes of Death; Mortality, Infant