In an exercise to demonstrate rounding bias in measurement, my students generated data that showed predominance of zeros and fives as the last digit of measured lengths. Chi-square text showed that the distribution differed significantly from the expected.
Some of the last digit numbers were very close to expected (i.e., the number 1 constituted 9.8% of all last digits vs. 10%). Others were not (e.g., 6 = only 6.7% of last digits). But most were close to 10%. My question: Is there a post-hoc test to determine _which_ of the measurements (here the last digits) was sufficiently different from the expected 10% expected to call them "significantly different than expected?" George P. Kraemer Leff Professor of Environmental Studies and Biology Chair, Environmental Studies Program Associate Dean, School of Natural and Social Sciences Purchase College 914-251-6640 (o) Website<http://www.purchase.edu/Departments/AcademicPrograms/faculty/GeorgeKraemer/GeorgeKraemer.aspx>
