You are not currently logged in.
Access your personal account or get JSTOR access through your library or other institution:
If You Use a Screen ReaderThis content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
An Approximation to a Multiple-Comparison Test
L. L. Eberhardt
Vol. 1968, No. 2 (Jun. 5, 1968), pp. 314-319
Stable URL: http://www.jstor.org/stable/1441758
Page Count: 6
Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Preview not available
A quick graphical means of approximating a standard multiple-comparison test amounts to using upper and lower limits which are multiples of the standard error, just as in the Dice-Leraas method, but using the square root of the number of means being considered as multiplier rather than "t" values or 2 as multiplier. Non-overlap of limits is then considered as evidence of significantly different means. For large numbers of means, the test becomes quite conservative and it is suggested that inner limits 9/10 the size of the above will be preferable, particularly when sample sizes are nearly equal for a given pair of means.