Voltolini wrote:
>
> Hi,
>
> I was reading a definition of "experiment" in science to be used in a
> lecture and the use of treatments and controls are an important feature of
> an experiment but.... my doubt is... is it possible to plan an experiment
> without a control and call this as an "experiment" ?
>
> For example, in a polluted river basin there is a gradient of contamination
> and someone are interested in to compare the fish diversity in ten rivers of
> this basin. Then, the "pollution level" are the treatment (with ten levels)
> but if there is not a clean river in the basin, I cannot use a control !
>
> Is this an experiment anyway ?
It's not an experiment, but not for the reason you are thinking. The reason
it's not an experiment is because there has been no assignment of exposure or
intervention to the units being investigated; rather, levels of the exposure
(pollution) occur naturally among the groups being investigated (fish
populations of each river). Hence, your study is not experimental, but rather,
observational.
As to the subject of control groups: there is no rule that says a control group
has to have a zero level of the exposure. You can use, for instance, the river
with the lowest pollution level as your control group ("reference group" might
be a better name) and compare the effects of higher levels of pollution to that
group. If you can quantify pollution levels, you could also model the effect of
pollution level as a continuous variable in such a way that the lowest level of
pollution in the study would implicitly be the reference level.
The problem that you have in any observational study, however, is with drawing
causal inferences from the results. Any observed association between pollution
level and species diversity could at least logically be caused by any
characteristic of the rivers that differed among them and that was associated
with their pollution levels. If there are such extraneous differences, they
must be controlled in some manner in order to make valid inferences about the
effect of pollution on species diversity.
-Jay
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at
http://jse.stat.ncsu.edu/
=================================================================
