Scientific Method


 * "The Scientific Method can never verify truth, it can only show what is not true." - Jim Haines

The Scientific Method is the general procedure by which any one individual can conduct an impartial investigation into the unknown. The following steps are considered to be the general outline of a full experimental procedure. The following should be expanded upon and adapted to each investigation.

It should also be noted that there are two sub-fields of virtually every scientific discipline that each use about half of the method each. Theorists usually work with the top half, using observations and experimental evidence to generate explanations for physical phenomenon and then make predictions about objects displaying such physical phenomenon. Experimentalists usually work with the bottom half, taking predictions made by theories to design and execute experiments. After analyzing their data they are then able to test the results against the predictions made to check the validity of the given theory. This does not mean that Theorists never conduct experiments or check the validity of their own predictions, nor does it imply that experimentalists never theorize as to the nature of their observations, but it does in general define the primary focus of each sub group.

Step 1: Develop the Question
The first step in any scientific inquiry is the look at what you’re investigating, and determine what it is that you wish to know about it. This could be information returned to you from a previous set of experiments, or it could be a set of observations about an unknown phenomenon. Either way the first objective is to determine what specific question you are going to answer.

Examples:

Observations of the Sky leads to the question, "What Causes the Sky to turn Blue during the Day?"

Previous Investigation into DNA results in a composition of 4 nucleotides. This leads to the question, "What is the structure of these nucleotides in the DNA molecule?"

Step 2: Hypothesize
Once you have identified the question that you're seeking to answer the next step is to examine the data again to make an educated guess as to what you expect the answer of your question, and result of your experiment to be. The hypothesis must be an answer to the question asked and must lead to testable predictions. A good hypothesis can make all the difference.

Examples:

After asking "What Causes the Sky to turn Blue during the Day?" You look observe the sky and the world again and hypothesize that the Sky is reflecting the color of the oceans and lakes on earth.

After asking about the structure of the DNA molecule, several scientists hypothesize that the DNA Molecule has a helical structure.

Step 3: Prediction
Once you have your hypothesis you must then figure out what a consequence of that specific hypothesis would be that is also able to be tested. This can be one of the most difficult parts of the process since the prediction must be unique to your hypothesis that way if you your prediction is confirmed you know that your hypothesis is correct. If your prediction is too broad confirming it does little for confirming your hypothesis over any others that would also lead to the same predictions.

Example:

After Hypothesizing that the Waters are the source of the skies color the scientist would predict that the sky near a lake that was not blue would also be a different color.

After hypothesizing that DNA is a helical structure one would predict that the X-ray scan of a DNA Molecule would be X shaped for any rotation of the molecule.

Step 4: Experimentation
Now that you've made a prediction, your next step is to create an experiment to test your claim. The Experiment should be designed to control any variables that aren't directly related to the claim being tested, allowing for clear results to be obtained that depend only on the tested variables.

Example:

Claiming that a Green Lake should be under a Green Sky our scientists looks up the locations of oddly colored lakes and travels to one. The scientists plans to observe the skies near this non-blue lake to see if the sky is still blue or if it matches the color of the lake. To minimize the error he plans to invite along several of his friends to observe with him, and to prevent error he's told them all to write down what they observe before discussing it.

Claiming that an X-shaped diffraction pattern under an X-ray scan, the scientists plan to isolate several molecules of DNA and take X-ray photographs of them to see what shapes develop.

Step 5: Analysis
Once the Experiment is conducted and the data is properly collected the next task to analyze that data. What did the experiment tell you about the variable you tested? Does that result agree or disagree with your hypothesis? Arguably more important than either of these two questions is: What are my sources of error and how big are they.

Most people think of error as something that's bad and isn't something you should tell people about, but the error is the only way to determine if your results are actually meaningful. If for instance you expect the value of a particular measurement to be 3, and you get an answer of 5, then you can say absolutely nothing. However if I say my answer was 5 with an error of plus-or-minus 2 then I know that the expected value could have been the actual value and I got a 5 simply because of the error I had. On the other hand if I measure 5 with an error of plus-or-minus 0.5 then I know something was wrong with the initial hypothesis and I can try and develop a new one.

You must use the results of the experiment as well as the possible error to draw conclusions about your prediction and hypothesis and be able to defend that conclusion with the data.

Example:

Our explorer arrives at Lake Hillier, a Pink Lake in Australia, along with several of his friends and colleagues to observe the sky. They all look up, take their time and write down the color that they see. The results come back and all 8 of them see a blue sky. Out of 8 people 6 are males and 2 are females. With an 8% colorblindness rate among men and 0.5% among women, there is only a 0.00000128% - 0.000328% chance that more than half of them are colorblind. Thus the error here will be very small and our scientist can conclude that the sky is in fact blue over the pink lake and his hypothesis was incorrect.

Our X-ray team similarly takes several X-ray photographs and looks at the results. They find that the vast majority of the images have X shapes in them with the remainder being difficult to make out any shapes. Their error calculations are much more intensive looking at the probability that the DNA molecule would scatter an X-ray or absorb it, the image resolution of their film, the likely hood that the X-ray would destroy the DNA molecule and punch straight through solid structure, but in the end they are able to determine their error is small enough to allow them to say that the DNA is infect helically shaped.

Step 6: Publication and Replication
It's great that we've come this far. We started with something we observed or learned from a previous experiment, asked ourselves something about the nature of that knowledge, developed a hypothesis to answer this question, used that to create an observable prediction of an experiment, designed and carried out an experiment, analyzed the results and determined the veracity of our initial hypothesis, but all of that means nothing without this step. We must write down everything we did, how we did it, and what our results were, how we analyzed those results and what conclusions they allowed us to draw, and then we have to share that information with other scientists.

These other scientists now have the ability to conduct the exact same experiments we did to check if the results agree. This is important because no one and nothing is ever perfect, and a scientific result that can't be replicated is no result at all. These scientists can also take the results of our experiment to now form their own questions, starting the whole process all over again and allowing knowledge to move forward one experiment at a time.

Examples:

Our Sky Scientist might write a brief paper explaining that he believed the blue color in the sky was a reflection of the blue color of the lakes. He would explain how he conducted the experiment and how he controlled for the other variables like peer pressure to conform among the group. Then he would list the results of the experiment and the sources and size of the error before concluding that the Sky's color was not a reflection of the waters. This would allow someone else to conduct a similar experiment to verify that, while statistically unlikely, our first scientists group wasn't completely colorblind. It would also allow for a new scientist to propose that the sunlight hitting the atmosphere scatters at different angles depending on the wavelength of the light with blue light scattering more than red hence the sky looking blue.

Our DNA experimentalists would also list their hypothesis of a helix shaped molecule and the prediction this made. They would explain how they took the X-ray photos and then the results of those photos. Other scientists can now scrutinize their procedure to make sure that they didn't do something that would mess up the results of the experiment and some of them could use the helical structure and start trying to figure out how the DNA unwinds to communicate it's instructions to the cell.