다국어인구학사전, 두 번째 통합본, 한국어판

주 의 보고계신 다국어인구학사전은 현재 작업 중에 있습니다. 자세한 내용은 토론부분에서 확인해주세요.

150

When the movement of a demographic variable in time is considered, a demographic time series¹ is obtained. It is sometimes possible to decompose a time series into a trend² around which there are fluctuations³, variations³, or deviations³ (141-2). Where such fluctuations tend to recur after certain periods, usually several years, they are called cyclical fluctuations⁴ or, more generally, period fluctuations⁴. In demography the most common period for compiling data is a year, and the fluctuations in sub-periods of a year are called seasonal fluctuations⁵. The fluctuations that remain after trend, cyclical, and seasonal fluctuations have been eliminated are called irregular fluctuations⁶. They may be due to exceptional factors such as wartime mobilization, or sometimes they are chance fluctuations⁷ or random fluctuations⁷. 3, In a general sense the term variation may be used to describe change in any value or set of values for a variable.

• 4. Periodic, adj. - period, n. - periodicity, n. cyclical, adj. - cycle, n.
• 7. Random, adj.: under the influence of chance (cf. 161-1).

151

It is occasionally desirable to replace a series of figures by another series that shows greater regularity. This process is known as graduation¹ or smoothing¹, and it generally consists of passing a smooth curve through a number of points in the time series or other series, such as the number of persons distributed by reported age. If a free-hand curve is drawn the process is called graphic graduation². When analytical mathematical methods are used, this is called curve fitting³. A mathematical curve is fitted to the data, possibly by the method of least squares⁴, which minimizes the sum of the squares of the differences between the original and the graduated series. Other methods include moving averages⁵ or involve the use of the calculus of finite differences⁶. Some of these procedures may be used for interpolation⁷, the estimation of values of the series at points intermediate between given values, or for extrapolation⁸, the estimation of values outside of the range for which it was given.

• 1. Graduation, n. - graduate, v. - graduated, adj. Smoothing, n. - smooth, v. - smoothed, adj.
• 7. Interpolation, n. - interpolate, v. - interpolated, adj.
• 8. Extrapolation, n. - extrapolate, v. - extrapolated, adj.

152

It is often necessary to graduate distributions to correct the tendency of people to give their replies in round numbers¹. Heaping² or digit preference² is particularly frequent in age distributions and reflects a tendency for people to state their ages in numbers ending with 0, 5, or other preferred digits. Age heaping³ is sometimes measured with indices of age preference⁴. Age data must often be corrected for other forms of age misreporting⁵ or age reporting bias⁵.

153

The numerical values of demographic functions are generally listed in tables¹, such as life tables (431-1), fertility tables (634-1), or nuptiality tables (522-1). A distinction is usually made between calendar-year tables² or period tables² which are based upon observations collected during a limited period of time, and cohort tables³ or generation tables³ which deal with the experience of a cohort throughout its lifetime. A multiple decrement table⁴ illustrates the simultaneous effects of several non-renewable events, such as the effects of first marriage and death on the single population. The most used are double decrement tables⁴. 틀:NewTextTerm provide numerical values of demographic functions, like survival functions (431-6) for example, which can be used directly for population forecast (cf. 720-2). When a population is classified in two or more categories according to age, like economic status (women in the labor force or out of the labor force, for example), marital statuses, regions etc. and when continuous flows between categories are possible over time even if the individual state can usually be measured only at discrete times (waves of a longitudinal study, queries to population registers etc.), 틀:NewTextTerm or 틀:NewTextTerm are more and more developed and used.

154

Where insufficient data exist to establish the value of a given variable accurately, attempts may be made to estimate¹ this value. The process is called estimation² and the resulting value an estimate³. Where data are practically non-existent a conjecture⁴ may sometimes be made to establish the variable’s order of magnitude⁵ .

155

Methods of graphic representation¹ or diagrammatic representation¹ may be used to illustrate an argument. The data are represented in a figure², graph², statistical chart³ or map³. A schematic representation of the relationships between variables is often called a diagram⁴, for example the Lexis Diagram (cf. 437). A graph in which one co-ordinate axis is graduated logarithmically and the other arithmetically is called a semi-logarithmic graph⁵, though such graphs are often inaccurately referred to as logarithmic graphs⁵. A true logarithmic graph⁶ has both axes graduated logarithmically and is sometimes referred to as a double logarithmic graph⁶. A frequency distribution may be represented graphically by frequency polygons⁷ obtained by joining points representing class frequencies with straight lines, by a histogram⁸, where class frequencies are represented by the area of a rectangle with the class interval as its base, by bar charts⁹, in which the class frequencies are proportionate to the length of a bar or by an ogive¹⁰ representing the cumulative frequency distribution.