- [Ron] - So why is it important to think about variables? Well, I
would argue that it's not only important to think about them, it's
also important to write them down. And the reason for this, is that we
often have judgements about what is important and what is not
important in an experiment. And sometimes we can make grave errors in
those assumptions.

- As an experimenter, it's important to keep track of what your
independent variables are and what your dependent variables are.
Because both of these are really important for setting up your
experiment, carrying it out, and also analyzing the data afterwards.
And, to keep them straight in my head, I often think of the phrase,
what is the effect of X on Y? The X is essentially your independent
variable. Whatever you yourself are varying or altering in the
experiment. And the Y is your dependent variable, what you're
tracking. An example specific from my project, using this phrase,
would be what is the effect of water availability on root patterning?
The water availability is the independent variable. Root patterning is
the dependent variable. You can think of controlled variables as
factors that you are consciously making sure do not change from one
sample to the next. For my experiments, I was making sure that the
light conditions of the experiment were kept constant. So all the
plants were experiencing the same lighting environment. That would be
a controlled variable.

- When you do in vivo work, there is also variables that are
associated with the animals, like the age of the animals. When we
injected the wild type cells and the HER3 knockout cells, we made sure
that all the mice that we were injecting where the same age. This is
critical because if you inject one of the cells in mice that is
slightly older, that can affect the tumor growth in these mice. 

[Neil]
- A confounding variable is something that will differ from sample to
sample, in a way that you are, for whatever reason, not keeping track
of or you're not keeping consistent. And the presence of this
confounding variable can also make your experiment more difficult to
interpret.

- In our laboratory we've done many experiments in the past using RNAi
knockdown. And, we found variability in the level of gene knockdown
even though we try to keep all the conditions exactly the same. And
after quite a bit of time we traced down the quote unquote confounding
variable and it turned out to be a very specific lot of serum. And we
discovered that some lots of serum from this one company allowed
effective gene knockdown by RNAi and other lots of serum didn't. So, I
think in these situations, it's really critical to then do experiments
where you can test one variable at a time. If your variables are
constantly changing and multiple variables are changing from one
experiment to the next, it just becomes very, very confusing to
compare results from different experiments and interpret them. So,
being methodical and really thinking through the variables, changing
variable one at a time, so that the results of an experiment are
clearly interpretable, also allows you to navigate your project in the
most linear and successful path.