Round numbers

Abstract

Numbers are used for exact and approximative estimations. The numbers used in approximative expressions are typically so-called round numbers, such as 10, 20, 25, 30, 40, 50, 100, 1,000, and such numbers are also very frequent in texts. This article presents evidence that some numbers are rounder than others and discusses how the roundness of a number can be derived from its contents of the base number of the numeral system of the culture. A formula for deriving the roundness of a number is suggested, and some evidence that intuitions about roundness vary between vigesimal and decimal cultures as predicted by the formula is given.

The language user as an arithmetician

“…In such expressions not all combinations of numerals are possible. This is in agreement with observations by Channell (1980) and Sigurd (1988), the only researchers who have paid some attention to estimation pairs. In accordance with their judgments, we claim that cases like "two or seven books" and "twelve or ten journals" are not acceptable as expressions in Dutch or any other language.…”

“…And, indeed, all cases in the corpora contain at least one number that is "round" in her sense. The same applies to all 45 examples presented in Sigurd (1988). Channell's concept of "roundness" and her generalization, however, are not correct.…”

Do Consumers Perceive Precise Prices to be Lower than Round Prices? Evidence from Laboratory and Market Data

“…For example, $450,120 would be considered round, while $450,127 would be considered precise. This measure is consistent with the definition of precision offered by Sigurd (1988). However, as we discussed above, roundedness is likely to be context-specific.…”

“…First, following Sigurd (1988), we consider multiples of 10 as round numbers and other numbers as precise numbers.…”

On Round Numbers: Pragmatic Aspects of Numerical Expressions

“…Roundness (defined as number frequency in an approximation context) proves to be related to three arithmetical properties: 'lO-ness', '2-ness', and '5-ness'. In predicting the frequency of numbers irrespective of their context '2Y2-ness' should be added to these factors, as is suggested in the work of Sigurd (1988). The role of the four number characteristics found in this study can be explained by the preference of the language user for using base numbers, and for doubling and halving quantities.…”