In the last set of videos,
we have looked at the three different types of experimental designs,
namely causal research, descriptive research and exploratory research.
Now remember when we talked about causal research, we mentioned that one
of the basic ways you can test the evidence of causality is experiments.
So in this video,
we are going to cover experimental designs and what we mean by experiments,
what are the different elements of experiments and then take it from there.
So what is an experiment?
Experiment is usually the manipulation and control of one or
more variables and based on the effects of those variables, observed outcomes.
And as I mentioned in the beginning,
this experiment provides evidence of causal research.
Now in terms of experimentation, there are basically,
sudden elements to the entire experimentation process.
First one is, manipulating the treatment.
That is, once you have certain variables of interest, you change the different
levels of the variables, and see what are the effects of those variables.
Of course, you cannot check all the effects of the variables simultaneously,
so you have to control for certain variables, and this is the control finally
based on the effects of the manipulation and the control you observe the outcomes.
In an experimental scenario,
there are certain variables which are important to remember.
The first one is called the dependant variable.
This is the variable which measures the effect of the manipulation, and
this is often called the outcome variable.
The next set of variables are the independent variables.
These are the variables which are of interest.
These are the variables that are manipulated.
So this is often called the treatment variable, and
these are basically the elements which the researcher wants to explore.
And these are usually, definitely manipulated by the researchers themselves.
The third set of variables are called control variables.
These are control variables which are measured by the researcher.
They are not intended to measure the effects,
but they are intended to keep the other conditions constant.
The control values are not allowed to vary freely or
systematically with the independent variables.
The last set of variables are called extraneous variable.
These variables are something which cannot be controlled by the researcher.
These things happen when you're running an experiment, there are certain factors,
which are beyond your control, you have to accept that.
And so, this cannot be measured even by the researcher.
Sometimes what happens of course is because of these extraneous variables,
the effect of the treatment variable are biased.
The next important element of this experimentation process or
experimental designs is how do you measure the effect of a treatment?
There are two main ways you can do it.
First one is called wooden subject variation.
This is where the same subject, maybe that's an individual or a group,
is first measured in terms of the outcome variable.
Then the actual manipulation is implemented and
then measured once more after the manipulation is implemented.
So basically, you are capturing the difference of the after outcome
with the before outcome and the difference is the effect of the manipulation.
So this is called the wooden subject variation.
The between subject variation is that slightly but different concept.
This is where we are actually measuring two different individuals or
two different groups of people.
Our concern is not about what happened before and
how the manipulation changed the effect, but here our concern is,
are these two groups being affected by the same manipulation differently.
So here, the between subject variation is usually measured
as the after outcome of the two or more different groups.
And the difference is actually very critical to
understand whether two different individuals or
different segments react different to a particular manipulation.
So this is basically the summary of the experimental design process.
And we look forward to see you again in the next video.
Thank you. [MUSIC]