What are the steps in the scientific method?

5- or 7-Step Method
In the 5-step method which some students are asked to provide and typical of beginning science classes, we see the formation of the problem, a prediction that provides explanation, a procedure used to test the ideas, the observation of results in the procedure, and a conclusion based on all of the other steps. There is a bit more to the method, and it might look like this:

1. Ask and define the question.
2. Gather information and resources through observation.
3. Form a hypothesis.
4. Perform one or more experiments and collect and sort data.
5. Analyze the data.
6. Interpret the data and make conclusions that point to a hypothesis.
7. Formulate a "final" or "finished" hypothesis.

With the investigation concluded, the published results will be verified by other investigators, and the "tested" knowledge integrated into a larger whole of scientific information.

Refined Method
It is important to note that there is no one single scientific method. Every experiment is different and may or may not follow the exact steps; science is less structured than most realize. However, there are key elements of the experimental process that we can identify.

In experiments (and in everyday life), scientists (and non-scientists) use hypothetico-deductive reasoning, or "If...then logic" to identify and test problems and solutions.

The start of every experiment does not start with "asking a question." It actually starts just before that. If you think about it, you cannot ask a question without identifying a problem that you observe. You do not ask "how does an owl hunt at night?" without first observing that an owl successfully captures mice as a food source at night. So this is where we start -- observation.

Now we go to our question. You have just observed a natural phenomenon, and now comes the time to question why this is.

As is our nature, we set out to answer this question. But first we need a tentative solution to our problem/question in order to test this theory. This is called a hypothesis; an educated guess. It is important that this hypothesis be able to test in an experiment. In other words, your hypothesis cannot be "because ghosts are playing tricks", because this is untestable and outside the realm of science.

So we have our tentative answer to our problem/question, and now we need to test this hypothesis. But usually we don't rush head-on into a task without knowing what our result should be. Therefore, we make a prediction, which will explain our results.

We have our hypothesis that we're going to test, and we have our predicted result should the hypothesis be true. Finally, we get to test and perform the experiment.

If this test supports the hypothesis, then additional predictions may be made and another test is performed. If the test does not support the hypothesis, then revision of the hypothesis is needed and a retest is performed.
 

Application of hypothetico-deductive reasoning

Observation: My flashlight doesn't work
Question: What's wrong with my flashlight?
Hypothesis: The flashlight's batteries are dead.
Prediction: If this hypothesis is correct
Experiment: and I replace the batteries with new ones,
Predicted Result: then the flashlight should work.