www.sciencebuddies.org
There is confusion sometimes as to what “science” is. For something to be science, the “scientific method” needs to be applied to it. (The diagram shows the different steps involved in this). This is a brief overview of it, if you want to delve into the philosophical arguments and so on, a search engine is your friend.
So first we should have a definition of what “science” is:
”an area of knowledge that is an object of study, particularly as concerning general laws; usually as obtained through specific rational methods.” (neurobioecthics.org)
WordNet: “a method of investigation involving observation and theory to test scientific hypotheses”.
Science does not busy itself with finding an absolute truth but rather defining a way of thinking. The idea is to continually test hypothesis and theories to modify, replace, or reject them if they are proved to be plain wrong.
Pretty much everyone agrees that there needs to be a method to make a rational assessment and reach a scientific conclusion.
Here we’ll look at what each step of the scientific method involves:
1 - Ask a question:
You see something interesting and you wonder if…why…how…what…when…where…how…? Here begins the process of human inquiry. Once the question is asked you establish a goal which is what you hope to find out about or discover, and also a plan which helps you go on to step 2.
As a very basic SEO example we could say “Why isn’t my site in Google?”
2 - Do background research:
You need to investigate an extensive body of knowledge. This means that you grab anything you can get your hands on about it and read it, make notes,etc…This is so that you can avoid making the same mistakes as someone else. It is also to make sure you don’t reinvent the wheel.
Saying “Ah it’s probably because a bad site is linking to me” when someone else has established already that this is highly unlikely is going to waste time and lead take us down the wrong path.
Research is a skill though. You need to be able to distinguish between relevant and irrelevant concepts, and have an accurate interpretation of other people’s experiences. It’s important to also find out who the authorities are on the subject.
In the SEO example we would find out who the most respected SEO professionals and researchers are, read all the information they have on the subject, trawl through all those other sources of information, and then sort through them. Does this post directly deal with my problem? Have I read this and understood it accurately?
Werner von Braun said “Research is what I’m doing when I don’t know what I’m doing.”
That’s the whole point of doing it.
3 - Construct a hypothesis:
A hypothesis is the state of the knowledge we have before we have carried out any experiments. After all the reading we might find it plausible that the robots.txt may not be scripted properly.
For a hypothesis to be considered correct, it must be confirmed through repeated experimental tests. If it isn’t testable, it’s not a hypothesis. We must also not be too attached to it and accept that the hypothesis could be correct or incorrect.
Dr Carl Sagan said “There are many hypotheses in science which are wrong. That’s perfectly all right; they’re the aperture to finding out what’s right. Science is a self-correcting process. To be accepted, new ideas must survive the most rigorous standards of evidence and scrutiny.”
4 - Test with an experiment:
They say “Experiment is supreme”. This is where we say “The proof is in the pudding”. This is where you make your pudding if you like. An experiment is a controlled test (or investigation). Results need to be observed and recorded.
There is a tendency (and we’ve all been there) when during an experiment we go “Ah-Ha! It’s True!” – but not so fast. This data needs to be gathered, sorted and then analysed.
In our very basic example we might modify the robots.txt and then see if the site gets indexed properly.
I always liked Ernest Rutherford’s quote: “If your result needs a statistician then you should design a better experiment”
5 - Analyse the results:
What happens at this stage is verification. Has our question been answered? Analysis can serve as a starting point for new hypothesis. Our original one can be ruled out or modified, if it is clearly and repeatedly incompatible with experimental tests.
Also don’t ignore or rule out data which do not support the hypothesis because this may be your most valuable find.
If the site still isn’t indexed after changing the robots.txt, then something else is wrong. Back to constructing a hypothesis. If it does get indexed…the process has been successful!
Alfred North Whitehead said “It requires a very unusual mind to undertake the analysis of the obvious.”
It is supremely difficult to analyse something where it is all painfully obvious because you have to put that aside. It’s amazing how often the obvious can be shown to be slightly flawed or plain incorrect.
In the Data analysis par for our little SEO example, we would looking at the results in Google, seeing how many pages are in, if any are missing and so on. If some are missing…new question!
6 - Report the results:
It is crucial to feedback the results of your experiment to the community. This is so that your experiment(s) can be reproduced (especially if it’s theoretical or highly experimental work).
It’s good when discussion around your work takes place. Sometimes everyone will disagree with you, and as long as the criticism is insightful and offers other possibilities, it’s brilliant! It means that maybe someone can come up with a better explanation, having carried out their own evaluation process. Sometimes lots of people will agree and this is also useful for everyone.
N.B: When a hypothesis is found to be wrong, it is just as valuable as finding that one is right. It stops people going down the wrong path.
In science there are very strict rules as to how you carry out evaluation, analysis and so on. Usually you share all of that with the community. This ensures that everyone re-evaluates in the same way. This way everyone is on the same page.
In our SEO example someone else may have carried out a check on your site and looked at your results and said “Actually, I think there’s something wrong with…” or everyone may agree and everyone has learnt that when something like this happens, a possible answer is yours.
Aaron Miles said “It is junk science. There’s no peer review. It’s jumping from one conclusion to the next.”
This why papers get published in academia and research because it’s difficult to claim something if it hasn’t been verified by an expert community. Open Source software also get rigorous peer review by all the other developers and users for example.
If everyone finds your hypothesis to be correct, it becomes a “Theory” but it must be flexible enough to be modified if new data or evidence is introduced. It will also need to applied to unrelated facts and new relationships to see how it performs.
If it is consistently right, experiment after experiment, and that it stands the test of time, it’s “Law“!
There is a song about it by Greg Crowther from the University of Washington, here’s a snippet:
“Information all around –
Some is bad, and some is sound.
How can I decide which statements to accept?
There’s a logical recourse:
Locate each primary source,
So conflicting sets of rumors can be checked”
One part of the scientific method which is often overlooked is the element of prediction. When you have shown that an experiment is consistent with your hypothesis. Can you use your hypothesis to show that predict a new phenomena? If the results of the prediction turn out to agree with further experiment this adds even more weight to your hypothesis.
We see this in many areas of science. Notably, in the periodic table. Arranging the elements in order of atomic number and by their properties allowed new elements to be predicted before they were discovered.
Dirac theorised the existence of the positron coming from a sea of negative energy. Much of Theoretical physics is about using exotic mathematical ideas to model the universe and predict the consequence of these ideas. Experimental physicists can perform the experiments to show the reality.
The SEO equivalent of prediction from results would be to perform an experiment and follow the logical consequence of this and then perform another experiment to see if the consequence was also true.
Brilliant input Ned, thanks! You’re quite right, prediction is very valuable.
I’m taking a “theories of personality” class right now and this ties in my scholastic work with my day job. Thanks for the brain boost!
Thanks for the outline and process. I have been looking for information like this so I can do a better job helping my clients. I have bookmarked this site and will come back often. Thanks, Zack
The diagram has it right; the scientific method consists of imagining a solution then constructing tests to see if reality confirms the idea. Personally, I’d love to see all the pseudo-science clutter cleared out, so that more of the world’s population could approach life from a realistic perspective – giving them a better chance of enjoying and improving their lives.