The Scientific Method:
A Brand New Story in the Making
“Beginning
Endlessly” from the album 20TEN (2010)
Have you ever wanted to conduct a
psychological study of your own? I knew it! The field of psychology,
like other sciences uses the scientific method when conducting
research and it bases the results from research on empirical evidence
(information obtained by observation or experimentation). The five
basic steps of the scientific method are as follows:
- Create a hypothesis (question) that can be tested
- Design a study and collect data
- Analyze results and reach conclusions
- Share the results with the scientific community
- Replicate the resultsEvery new experiment is A brand new story in the making, the scientific process is always beginning endlessly.
Pick a problem or question to
research
Why should you be satisfied with
just heaven and earth? Every
scientific inquiry begins with a question. From the beginning of time
man has asked questions that lead to scientific discoveries. Mankind
has not been satisfied with
just heaven and earth but
there are so many
unanswered questions out there that need to be answered. Three common
methods psychologists use to uncover a question to investigate are:
find an everyday problem (What is the best way to deal with road
rage?), investigate a societal belief (Are males more sexually active
than females?) or research already published psychology literature to
find a question that needs further examination.
Define your variables
Once you have your question picked out
you have to define your variables. A variable is anything that could
influence the results of your study. When doing experiments that
involve humans there are many that have to be considered (mood,
hunger, design of laboratory, etc.). Inside all of us are minerals
and chemicals of space, the
complexity of the human body makes experimenting accurately very
challenging. When defining the variables you are using you
have to describe how they are measured in the context of your study.
For instance if you were conducting a study on how Prince's music
affects a female's willingness to have sex you would need to
concretely define what you mean by Prince's music, you might define
it as: having listened to five sexually explicit songs by him
(Scandalous, Insatiable, Sexy Dancer, Do Me, Baby and Shh).
Willingness to have sex could be defined as: after listening to the
music, increased blood flow to the clitoris. It is very important to
define your variables so that they remain constant between all
participants.This is especially true for the independent variable
(IV). The IV is the thing that is being manipulated so that we can
study it, in this experiment Prince's music is the IV. The dependent
variable (DV) is the willingness to have sex.
Hypothesis
Humanity
is naturally curious, we constantly ponder the sublime questions. Are
we made of stardust?
If you look around there's so much more to the universe, Maybe every shining star, is just another part [of us]. As well as more vulgar one's like in our example above. Once we have our question we next formulate a hypothesis. A hypothesis is a concise statement that states what the point of the experiment is, for example: After listening to Prince's music women will be more willing to engage in sexual activity than women who do not. Once we have or hypothesis we then construct a null hypothesis. A null hypothesis is required in any experiment because it helps to show if a study is clinically significant or not. A null hypothesis is the prediction that one variable will have no association to the other variable. For our purposes that would mean that Prince's music has no affect on female whether a female will want to have sex or not. At the end of the experiment the researcher has to look at the data, analyze it and see which is true the hypothesis or the null hypothesis.
If you look around there's so much more to the universe, Maybe every shining star, is just another part [of us]. As well as more vulgar one's like in our example above. Once we have our question we next formulate a hypothesis. A hypothesis is a concise statement that states what the point of the experiment is, for example: After listening to Prince's music women will be more willing to engage in sexual activity than women who do not. Once we have or hypothesis we then construct a null hypothesis. A null hypothesis is required in any experiment because it helps to show if a study is clinically significant or not. A null hypothesis is the prediction that one variable will have no association to the other variable. For our purposes that would mean that Prince's music has no affect on female whether a female will want to have sex or not. At the end of the experiment the researcher has to look at the data, analyze it and see which is true the hypothesis or the null hypothesis.
Do Your Research
History is full of praise for
another's history. This is so
true for this step in the process. During your research you will
praise those who came before you because most of the work you will
find has been done already. A great place to start is The
Educational Resources Information Center
(ERIC) this website has over a million references that researchers
commonly use. As you
research take notes and
gather resources that can assist you in your experiment. You can use
books, peer reviewed articles and websites. I would avoid any Dr. Oz
related resources though. In our example we might look for research
on music's affect on the female brain, sexuality and music, pop
music's affect on cultural attitudes towards sex. Stuff like that.
Make sure you start creating a bibliography of all the sources you
used because a study without any sources might be okay with Dr. Phil
but it will not be accepted by the scientific community. How many
sources should you use? The more the better.
Choose the type of experiment you
want to do
There are many different experiment formats to follow but we will
talk about the three most common. First up is the True Experiment
Format. This is the one most of us are familiar with. In this
experiment there are two groups the control group (the group of
ladies who will not be hearing any Prince music instead they will
hear non-sexual classical music) and the experimental group (this
group will get the sexy Prince music). All participants will be
randomly assigned to one of these groups. After the IV has been
administered both groups will have the DV (blood flow down there)
measured to see if there was a noticeable difference between the two
groups. The second experiment format is the Pre-experimental one.
Here there is no control group instead one group is administered the
IV and then are studied. For example each participant is given a test
before the experiment and then after to see if the IV had a
significant affect on them. Case studies are a popular way to conduct
a Pre-experimental study. The last format we will look at is
the Qusai-experiment format. This format does include a
control group but randomization does not take place.
Standardization
of process
Left alone with our devices,
whatever will be will be. This
is so true, if we don't have rigid implementation in studies then
reliable results cannot be obtained. It's best to have someone help
you with this step so that you get another perspective to ensure that
you have accounted for as many variables as possible. Standardization
means you determine
exactly how you are going to conduct the experiment. In other words,
construct your experiment so that others can conduct it the same way
you did because the last step in the Scientific Method is
replication. If the results cannot be replicated they are not
reliable. When testing the participants all the conditions have to be
the same for all participants in the control group (if you are using
one) and the experimental group. So all the women who are in the
experimental group will all be exposed to the IV in the same way
(same song order, temperature in the room, amount of rest, etc.). The
more similar you can make the experiment the greater chance you have
to have an accurate study because you didn't allow outside variables
in to distort or confound the results. Of course you can never get
100% control over all the variables that's why you will never see a
study that is 100% accurate.
Select
Your Lab Rats
Two
main ways of doing this are a simple random sample (randomly choosing
people from the population) and a stratified random sample (a random
sample that draws from subsets in the population, subsets could
include geographic location, age, sex, race, or
socioeconomic status). Once
you select the group to experiment on you should ensure that each
candidate is a good pick for the experiment. For instance you
wouldn't want someone who only listens to Christian music, has a
strong aversion towards Prince or is a Prince fanatic to be part of
the study because they could skew the results. You might administer a
questionnaire beforehand to help eliminate these sorts of
participants.
Conduct
the experiment and collect the data
Safety
first, before you start experimenting on anyone you have to make sure
that your experiment is ethical. The famous Stanford Prison
Experiment today would be considered unethical if you are not
familiar with that study please visit here:
https://en.wikipedia.org/wiki/Stanford_prison_experiment.
It was a fascinating study that will make you lose faith in humanity
(think Abu
Ghraib).
Whatever institution you are conducting the experiment for will
require you to submit the details of your study to their
Institutional Review Board to make sure you are
ethically sound. Next you would obtain written consent from your
participants letting them know of the process and the potential risks
they could encounter. Now you are free to start experimenting and
collecting data! If you are like me and have a strong aversion to
numbers than use the Statistical
Package for the Social Sciences (SPSS), it is a computer program that
will do all the computations for you as you collect your data. The
internet is great!
Analyze the data
Did the data
support your hypothesis or the null hypothesis? This is the tricky
part of the scientific method you have to determine did Prince's
music truly cause increased blood flow to the clitoris or was it
something else? Or was there no noticeable difference? This is where
a Type I error and Type II error can derail an experiment. Type I
error is when the researcher rejects the null hypothesis when they
should accept it. A Type II error is when you accept the null
hypothesis when you should reject it.
Peer Review
You've
made it to the last step. You write up your results using the APA
(American Psychological Association) professional journal format
which has the following sections: title page, abstract,
introduction, method, results, discussion, references, tables and
figures. Finally you submit it to the board of a professional journal
so that they can review it and determine if your experiment was
executed correctly. If so congratulations you now have added to the
scientific process! Now wasn't that fun?
Reference
Mitchoff, Kate Houston. "Scientific
method." Teacher Librarian 34.4 (2007): 51.