Scientific Method - Part 1

Introduction

So these days we seem to have problems with how the general population views the scientific method.  There are many misconceptions with how science works and this leads to people who deny climate change, creationists and people who put their children's lives and the lives of others at risk by refusing vaccinations.

So how does science work?  Why do these people exist? Is there really a scientific conspiracy to keep churning out the same research and hide the truth from us?  In case you were wondering, the answer to the last one is no.

Paradigm

First, let's look at what is meant by the term "paradigm." The idea of paradigms was first introduced by Thomas Kuhn; he suggested that a paradigm was “the practices that define a scientific discipline at certain point in time.” This is more than "just a theory" (we will look at the flaws of this phrase shortly) but instead the whole school of thought. It is the direction in which research goes and how it should be performed; it contains all the scientific views in a particular field of study.

Now paradigms can change, these changes are usually called a "paradigm shift." Originally Kuhn proposed that these would be a change so great that it would take science in a whole new direction. Take Newtonian physics as an example.

Newton came up with the laws of motion that we still learn in school today while still studying at university.  Then after some correspondence with Hooke, Newton started working on the orbits of planets and why they display retrograde motion in our sky.  In order to complete his work Newton needed to invent integral and differential calculus; a whole new branch of mathematics!  Now he did all this, and lots more, when he was 25! I know, right?

However with the discovery of subatomic structure this school of thought no longer held up. Einstein managed to apply a huge force to this steam train of a paradigm and change its momentum in a very short amount of time. Now research into mechanics is mainly based around quantum mechanics and Einstein's theories. This is a new paradigm.

It should be noted here that a paradigm shift is not just saying why everyone before you was wrong. Newton's laws held up for a very long time, so Einstein needed to explain why. Very rarely does a shift in the paradigm completely negate all that came before it, instead, the new paradigm will be an adaptation: encompassing the laws from the previous paradigm but accounting for their faults.

It's only a theory

I need to talk about theories now. There is a fundamental misunderstanding about what the word theory means in a scientific context. This is caused by how the word is used in everyday life:

"The way I had it figured, Toruk's the baddest cat in the sky. Nothing attacks him. So why would he ever look up? That was just a theory." Jake Sully, Avatar

This is not a theory, this is simply an idea: a thought that Sully has; nothing more than conjecture based on being attacked by it earlier in the film.
So what actually is a theory? Well it's far more than an idea or simple conjecture. A scientific theory is an explanation of observable phenomena that is based on a collection of facts that have been repeatedly confirmed by experiment. This means that a scientific theory holds a lot of weight. Meitner didn't just sit at her desk, glance up at her lab partner and say "I have a new theory! If an atomic nucleus breaks apart it gives off lots of energy and we could use that to generate heat."

She made observations of this heat being given off after she fired some neutrons at some uranium.

Her lab then tested this over and over to check if she was right.

She then worked to come up with an explanation that fully described all of what was happening and why. She did not change her observations or calculations simply to make it fit.

Then when other scientists heard of her hypothesis (yes, hypothesis; not theory) they tried the experiments too and tested the hypothesis repeatedly.  Not one experiment showed any flaw in Meitner's explanation.

Now she has a theory.

To put this in more relatable terms, if you look out your window and think to yourself that it is light outside because the sun is out then what you have is an hypothesis (If it is light outside then the sun must be out). You do not have a theory.
If you go outside and look up to check that the sun is out, then you have backed up your hypothesis with experimental evidence. Now if hundreds of people check over and over again that every time it is light outside that the sun is out then, and only then, you have a theory. If one person finds a time when it is light outside but the sun is not out (maybe a street lamp is on) then this destroys your whole theory and it must be adapted to account for these new observations and the whole process starts again.

A theory has been repeatedly put to the test and in each case the theory has successfully explained the results of that test.

I hope now you understand why I dislike the phrase "it's only a theory" which is thrown about so much by a certain group of people.
Try jumping out of a plane. If you think gravity is a theory, bring a parachute. If you think it's "only a theory," try the jump without one, you never know what could happen....

Hypothesis

I have mentioned the word hypothesis a few times in this post so far so I thought I should just clarify what it means and how it is different from a theory.
An hypothesis is a proposed explanation for a phenomena; it is usually phrased as a statement that attempts to explain the reasons for why things happen the way they do.
A good hypothesis should contain all of these features:

• Testability, i.e. it should be phrased in such a way that a test can be done to either support or refute the hypothesis
• Parsimony.  This means it should not make too many assumptions; Occam's Razor states that with two competing hypotheses that the one that requires the fewest assumptions should be selected.
• It should be able to explain more phenomena in the future
• It should "fit" to some degree with the current paradigm

There are exceptions to this list; otherwise we could never have a paradigm shift but it is a good general rule of thumb.

So this is an example of a good hypothesis:

If the rate of reaction depends on the temperature of the reagents then changing the temperature the reaction is carried out in will result in a different rate of reaction.

It is fairly simple, however it is testable and refutable.  A test could prove this wrong. It does not rely on many assumptions as all the ideas are backed by collision theory.

Here is another example of an hypothesis

Life as we know it today has been intelligently designed by a superior being

This is not testable; there is no way I can see to test this hypothesis and more importantly, no test that can show this hypothesis to be untrue.
It relies on a massive assumption (the existence of a superior being) and when you try to get it to explain further phenomena, such as fossils, it requires even more assumptions.

To be continued...

References

http://skepticcoffeebreak.tumblr.com/
http://explorable.com/
http://www.aaas.org/news/press_room/evolution/qanda.shtml