Log in

No account? Create an account

Previous Entry | Next Entry

The Scientific Method: A Primer

Among regular people there seems to be precious little understanding of what exactly the method is, and what it does and does not accomplish.  This ignorance about the process of science contributes to claims that scientists are just greedy a-holes exploiting the government for profit.  This attitude rises from a complete lack of exposure to real scientists and their way of being.  It is not fair to scientists.  Scientists, for the most part, are trying to figure out how the world works so that we can use that information to make our lives and our world better.  They are not politicians, they are curious people who sought education enough to know what questions to ask and how to test them.  They care passionately about making the world a better place.

The first step of DOING science is to ask a question about the world.  The question doesn't have to be complicated, it just has to reach into the unknown.  Once you have your question, it is a good idea to snoop around and see if anyone else has already answered it, or tried.  Learn everything you can about the variables that might influence the answer.  Once you've studied up on it, you're qualified to make a guess---a "theory" in science terms---as to what the answer might be, and why.  A true scientist knows that a theory is just a theory--it has to be tested repeatedly by people who agree and by people who disagree.  A true scientist is not heartbroken when the data shows that his theory was bunk.  That is useful information.  Time to come up with a new theory.

This testing is the experiment.  There can be many different ways to test any one theory.  The most useful experiments are often the simplest, changing only one variable between two groups of test subjects.  Scientist use many different methods to approach the same question, and this diversity adds richness to the picture painted by the results.  We might know that B follows A three quarters of the time, but until we know WHY they are correlated, and what other variables contribute to the correlation, we do not understand.  A--->B at a rate of 75% is enough to know that there is a connection, but it is not enough to say that A causes B.  We don't know that.  Something else could be causing it.  We take our results from that experiment, share them with the other scientific thinkers in the world, and update our theory if possible.  Usually an experiment brings up new questions, which indicate new possible experiments that need to be done to understand.

So science does NOT discover causality.  It discovers correlations.  Correlations can have multiple contributing variables so more experiments are needed.  Sometimes someone repeats the same experiment and gets the opposite result.  This is evidence that there was something operating in the system that was not being measured.  This is a sign that the original theory was based in deeper ignorance than perhaps we thought at first.  This is hard to admit, even for scientists.

Just because an experiment gets peer reviewed and published in an journal does not make it the truth.  There are many false conclusions that have been published.  Egostists who call themselves scientists publish more books than all the real scientists put together.  Real scientists tent to be intraverts who'd rather stay out of the limelight and just keep digging into these interesting questions.  Every experient needs to be repeated from a variety of angles before a result is accepted as Truth.

So there is a basic primer on the scientific method.  My area is mostly medicine, though I am fascinated by all science.  Medical science is more than double blind placebo controlled studies.  It includes the careful evaluation of population outcomes and biochemical mechanisms and every other factor that could influence the answer.  Science is a process of asking questions and trying to figure out if our theories about the answers are right or not.  A theory is just a theory.

Evolution, by the way, has been proven in so many ways by so many different experiments, that it is not a theory anymore.

Posts from This Journal by “science” Tag


Apr. 14th, 2016 05:48 am (UTC)
Yes, it is a good explanation but might be difficult for someone with no background. I would define independent, control, and dependent variables and show how they are manipulated in a couple of example experiments.

In the social sciences, the classical experimental method is often impossible, illegal, immoral and not practical. Experimentally asking about the effects of observing gun violence would require that some randomly chosen subjects be exposed to such violence. This is not ethical and the effects may not be reversible. So how do people investigate the effects of observing gun violence? They do it very poorly, i think. because they do not think experimentally when developing whatever alternative to an experiment they will use.
Apr. 14th, 2016 06:09 pm (UTC)
Yes, you are right, one could go much deeper into explaining how experiments are designed. And yes, a person without higher education could find this explanation challenging.

As for the social sciences, this is where Big Data shines. We can't in good faith subject people to violence or other known dangers, but we can collect data about all kinds of people and guns, and crunch it. Giant sample sizes and tracking many variables has helped us recognize correlations that no one had predicted. Hopefully this approach will yield useful information about gun violence as well. Unfortunately it depends on people being willing to report about themselves and their lives to a system that could be percieved as "Big Brother".

I have no answers, just more questions.
Apr. 15th, 2016 03:26 am (UTC)
That means you probably are a scientist.



Latest Month

October 2019


Page Summary

Powered by LiveJournal.com
Designed by chasethestars