If someone suggests that other breeds or females would behave differently
with the diets, we have no counter-argument.
We only have information about the breed of male rabbits we considered.
If we had taken a random sample of rabbits from several breeds, we introduce the variation inherent in those breeds.
Some breeds are smaller and more active than others, while others are larger and more sedate.
This variation makes it more
difficult for us to find a difference in weight gain due to diet if one exists.
However, if we do find a significant difference in
weight gain, we can say something about rabbits of different breeds, not just one special breed.
Similarly, if we used males
and females, our population of inference would be rabbits of either gender.
Through control, the experimenter attempts to accentuate or make as visible as possible the planned, systematic
variation between treatment groups, while at the same time reducing or removing as much chance-like variability as possible.
The smaller the population of inference, often, the greater the control we have.
A second approach to handling the chance-like variability is through randomization.
Clearly, it is not possible to
remove all chance-like variation through our methods of control.
Rabbits are still different, even if they are the same breed
and gender; some will grow faster than others, regardless of the diet.
Measurement error is always present even if the same
scales and technicians are used.
By randomly assigning the rabbits to the treatment groups, we will spread the chance-like variation among the
This adds to the variation in each group, but it removes the bias that would otherwise doom the
This random assignment of experimental unit to treatment group is essential for an experiment and distinguishes
it from an observational study.
In an observational study, the experimental units or subjects are not randomly assigned to the
As an example, if we looked at rabbits at Farm A who are presently being fed Diet A and rabbits at Farm
B who are presently being fed Diet B, we might find that the average weight gain for the rabbits at Farm A (on Diet A) had a
greater average weight gain.
However, the diet and farm are inextricably confounded.
We could never say that Diet A was
better than Diet B, since any weight gain could be a result of other aspects of rabbit life at the two farms.
If, in the end, we
want to say that one diet produces greater average weight gain in rabbits than the other, then randomization is essential.
Randomization plays other roles in our experiment.
We should place the rabbit cages in our designated locations
through a random process.
This keeps the lurking variables of heat, light, air flow, and other unknowable variables from
biasing the results.
These unknown variables can produce systematic, unplanned variation if randomization is not used.
effects of these variables, if any, is distributed to the two treatment groups by this randomization process.
This form of