The Social and Political Context of Population Forecasting

Keyfitz, N. (1984). The Social and Political Context of Population Forecasting. IIASA Working Paper. IIASA, Laxenburg, Austria: WP-84-003

Preview

Text WP-84-003.pdf Download (986kB) | Preview

Abstract

The subject of forecasting abounds in paradoxes. All statistical data refer to the past, yet for purposes of action only the future counts, and there is no necessary connection between past and future. Thus forecasting is on the one hand impossible, on the other hand indispensable. The difficulty is greater for population forecasts in that they are demanded for half a century or more ahead, where economic and other forecasts need cover no more than one or two years into the future.

Forecasts are needed for planning; plans are rarely for more than five years in the future; why does anyone want population forecasts for the next half century? Official forecasts are usually presented as alternative projections, among which the reader chooses; yet if the reader chooses that projection whose output numbers he prefers, then he might as well choose among a set of random numbers. Forecasts are often in error, yet there have been cases in which they were given with accuracy, and where they were disregarded: examples are declining school attendance and increasing pensioners.

Any discussion of the social and political context of anything gives the impression of denigrating that thing. Our intention here is to show the importance of statistics and of forecasts, despite errors and misunderstandings.

Actions (login required)

View Item