- When an observed difference in treatment effects is too large to reasonably have occurred
by chance, we say that the difference is statistically significant.
- If significant differences among treatments are found after running a comparative
randomized experiment, we conclude that the differences are CAUSED by the treatments.
- A mechanism is required to perform the process of randomization. E.g. table of random
numbers, or a computerized number generator.
Dr. Christina Hayes
Wilson 2-263
Department of Mathematical Sciences
Montana State University
Bozeman, MT 59717
phone: 406-994-6557
fax: 406-994-1789
christina.hayes@montana.edu
(Email will likely reach me faster than a phone call)