A STATISTICAL METHOD FOR MEASURING ‘MARGINAL UTILITY’ 159
The private account books, . . . are all full. . . . But it is chiefly
a want of method and completeness in this vast mass of information
 which prevents our employing it in the scientific investigations
 of the natural laws of Economics.” *
It has long seemed to me that just such records of consumption
as Jevons mentioned ought, on proper analysis, to yield a real
statistical measurement of this most elusive of magnitudes. If
this cannot yet be done for the individual, or even for the
individual family, it may be that it can be done for a “typical”
family, an imaginary family, consisting of a given number of
people and having a given income. That is to say, we may
possibly be able to make use of mass statistics somewhat as the
physicist measures the pressure of a gas without measuring the
impulses of individual molecules, though it is really the bombardment
 of these against the containing walls which really constitutes
 the gaseous pressure. We can often gauge a mass effect
when we cannot gauge its constituent parts. As my old master in
mathematical physics, J. Willard Gibbs, used to say, “The whole
is simpler than its parts.”
The records from which I hope we may succeed in distilling out
the desired psychological essense, the want-for-one-more, are (1)
retail prices, and (2) family budgets. Through such mass statistical
 measurements we may succeed in gauging average or
typical human emotions even better than any individual who
feels them, just as a clever editor, or advertiser, or salesman,
knows what the mass of people want better than any person in
that mass itself.

Specifications Re Budgets and Prices
The method consists, in a word, of so utilizing data of family
budgets and prices as to compare the wants of two typical families
of different incomes, in the same community, by using as a yardstick
 or criterion, a third typical family having identical tastes,
but differing in the amount of income, and living under a different
scale of prices for foods, rents, clothing and other items of
consumption.
Let us, then, imagine three typical workingmen’s families, each
consisting, say, of five people, the man himself, his wife, and
three typical children.
2 0p. Cit. pp. 10, 11.