Science! What is it and what is its purpose?
The word "Science" is derived from the Latin word Scientia meaning "Knowledge". Science is a systematic process that seeks knowledge about natural phenomena within the perceivable universe. In simpler terms, it's a method of rationalizing what we think of as real. Anything we can touch, taste, see, hear or smell can be tested with scientific methods. These methods change slightly from one field of research to the next, but all are subject to the same level of scrutiny. The scientific method can often be discouraging since the first rule of science is to prove why new ideas are wrong. All science is subject to Peer Review. In this way more experienced scientists can examine the experiments of another and look for mistakes and even bias in their methods. Additionally other scientist in the same field can attempt the same experiments to see if they produce different results. Many hypotheses are shot down like flaming airplanes because of these safeguards but it's a necessary function of science to keep our perception of the facts as clear and pure as possible.
What is the Scientific Method?
In general, the Scientific Method boils down to roughly five steps: Question, Hypothesis, Prediction, Test and Analysis. Sometimes there are more, sometimes less, and sometimes they're even in different orders but the point always remains the same; to learn about something that we don't currently understand. So let's take a moment to take a closer look at the scientific method.
- Question: Your question can be about anything that you don't currently understand. In the past the questions were as simple as "Where does the sun go at night?", "Why does water expand when it's frozen?" or "Why do things fall down?". So the question isn't limited to anything. Remember... if you can see it, touch it, smell it, taste it or hear it then it can be tested. As long as your question falls into one or more of those five criteria, then science can find an explanation to it. But this step isn't just about asking questions, it's also about educating yourself. In all probability you aren't the first person to ask what ever question you might have. So this step also involves researching the question to find out what information already exists on the subject. There's always a chance someone has already come to a completely reasonable answer, maybe even a definitive one. If that's the case, ask a different question, if not or you're not satisfied, do the tests again.
- Hypothesis: After you have a firm and solid question that is either without a satisfactory answer or without any answer at all, then you can make an educated guess based on the knowledge obtained when formulating your question. That's what this step is all about. Now that you know a thing or two about your question you can guess as to why your particular phenomenon behaves the way it does. It's not always that simple, in the case of statistics (such as in where populations are concerned), you would actually have to form two hypotheses, a Null Hypothesis and an Alternative Hypothesis. The Null Hypothesis is a guess as to if the Statistical Hypothesis proves to be false. The Alternative Hypothesis is the preferred result and is a guess that the Statistical Hypothesis is true. No matter what though, the Hypothesis must always be potentially falsifiable or it can't be effectively tested.
- Prediction: This step is basically guessing the likely outcome of the tests proposed in the Hypothesis. The evidence that the Hypothesis is true will be stronger if the outcome is not already known or less likely to be correct by coincidence. Additionally, if two hypotheses make the same prediction then proving the prediction to be true would not be evidence of either one over the other so it's best to make a prediction the distinguishes your hypothesis from other similar hypotheses.
- Test: In this step, one would preform experiments to test the hypothesis under real world conditions to see if reality behaves in the same manner predicted. Karl Popper advised scientists to design experiments to try and disprove hypotheses, even their own. If the experiments conflict with observations made in the real world, then confidence in the hypothesis will decrease. If they agree then confidence increases. Confidence in a hypothesis does not establish it as true or untrue either way and further testing would be needed. A hypothesis would need undeniable evidence to be considered either true or untrue.
- Analysis: Here one would gather the information yielded by the experiments and determine if the Hypothesis has been falsified. If it has then, a new question must be asked. If it has not been falsified but confidence is not high enough then new experiments may be tested for more predictions made by the hypothesis.
We're not done yet!
If you've done all that and think there's high confidence in your hypothesis you might think you're done... that the hypothesis is true and the argument is over. Unfortunately that's simply not the case. In science it's standard procedure to shoot down any new claims. You have to be ready to convince colleagues and peers of your research even if they don't want to be convinced. That's why all science will subscribe to the following three steps.
- Replication: All tests preformed must be repeatable and potentially falsifiable. If they aren't then the data suggests either error, or bias. Other scientists will often preform the same tests just to see if they can achieve the same results. Additionally they'll even preform different tests in an attempts to disprove the hypothesis altogether.
- Peer Review: Experts in the same field will anonymously review the work done for quality. They remain anonymous to prevent bias. Their opinion of the work also does not establish the results of the work as undeniably correct either, only that the experiments were sound. The reviewers may also request that other experiments be done to increase confidence in the hypothesis. If the work passes peer review then it may be published in a peer-reviewed scientific journal which would indicate quality in the work done.
- Disclosure: Every detail of the work done must be recorded precisely to allow others to replicate the work done and decrease their own bias. The data must be made available to other scientists so that the experiments may be repeated as close to the original as possible and even samples collected may need to be shared as well.