By trade, historians tend to be skeptics who prefer the specificity
of nitty-gritty facts to grand generalizations and fanciful speculations.
Such skepticism seems especially appropriate when dealing with claims
based on quantitative analysis. Although people often think of numeric
data as "hard" evidence, there is also a common perception
that experts can make numbers "say" anything they wish. As
the aphorism attributed to Mark Twain (among others) declares, "There
are three kinds of lies: lies, damned lies, and statistics." One
may be tempted to dismiss quantitative analysis because it seems obscure
and hence untrustworthy. Yet the information available in numeric form
can be too valuable for a good historian to ignore. Quantitative data
do not speak for themselves, but with a little coaxing they can sometimes
tell us things about the past that we cannot discover in "qualitative"
kinds of evidence.
The challenge for beginning historians is twofold: (1) to learn how
to pose good questions of available quantitative sources, including
both raw and aggregated data; and (2) to learn how to organize and "read"
the data yourself to answer the questions you have posed. If you do
not like mathematics, you probably will not become a heavy-duty quantitative
historian. But you can still use basic quantitative methods in your
research, and you can still become a critical reader of complicated
quantitative scholarship. There is a range of reasonable positions between
that of a true believer, on the one hand, and an anti-numeric nihilist,
on the other. The philosophy underlying this guide is that quantitative
history is too important to be left exclusively to the mathematically