As a statistican would say, "If in doubt, take logs." The world in general is probably lognormally rather than normally distributed, and thus many variables are better measured on a logarithmic scale (which is multiplicative) rather than an additive scale--just as graphics sometimes use ratio scales. Camera settings (f-stops, ISO numbers) are also multiplicative rather than additive.

For more, indeed a lot more, see Edward Tufte, Data Analysis for Politics and Policy, pages 108-131. This material looks at various interpretations of slopes (regression coefficients) for mixes of logarithmic and non-logarithmic scales, and why we might want to use log scales for many kinds of variables, possibly including The Lie Factor.

-- Edward Tufte

The distinguished statistician David Wallace often advises that "the second step in analyzing data is to take their antilog". See

Sex, Smoking and Life Insurance. Chance, 14(3), 42-45, 2001.

For two compelling examples of how well it works,

-- Howard Wainer (email)

See Edward Tufte,Data Analysis for Politics and Policy (1974), end of chapter 3 for plenty of material on log-log regresssions. Also check out an econometrics textbook, since log-log regressions are used to estimate elasticities in economics. Tell your reviewers to look at DAPP for political examples.

It is straightforward. The regression slopes in log-log space estimate percentage changes (that is, a 1% change in X is associated with a b% change in Y, just like elasticities) instead of unit changes in standard X Y space (a change of 1 unit in X is associated with a change of b units in Y).

-- Edward Tufte