Stevens: Subjective scaling data collected by S. S. Stevens
Description
This experiment was apparently done by S. S. Stevens and colleagues in March 1962,
although the exact reference is lost. 10 subjects were played tones at each of
5 loudnesses, presumably in random order. Subjects were asked to
draw a line on paper whose length matched the loudness of the tone. Each
subject repeated each loudness 3 times, for a total of 30 trials per subject.
The original data are lost; reported here is the mean of the 3 log-lengths for
each loudness, the sd of the three log-lengths, and the number of replications,
which is always 3.source
These data were obtained in the early 1970s from the data library in the
Harvard University Statistics Department.Details
This is a classic example of a psychophysics experiment pioneered by S. S.
Stevens. The basic idea is that the phychological response y to a physical
stimulus x should be proportional to x to a power. Since both the response
and the loudness are already in log-scale, linear fits should be expected.References
Stevens, S. S. (1966). A metric for social consensus, Science, 151,
530-541, http://www.jstor.org/stable/1717034