Modern science has arisen from strenuous efforts over the past few centuries (especially since Galileo, Bacon, etc) to ensure that our evidence (data), and the conclusions we draw from it, are watertight. This has evolved into what we now call 'scientific method', which is based on the use of repeatable experiments and objective reasoning.

A theory is not held to be true because some important people say it is, or because its proponents talk louder and longer than anyone else, or because it promises to keep my skin soft and smooth forever...

·         Objective Experimentation and Observation.

Science depends heavily on the repeatability of experiments, and on their giving consistent (e.g. almost identical) results. This repeatability hinges on objective comparison of observations of different researchers studying the phenomenon.

Objectivity indicates the desire to observe things as they are, without manipulating the observational results to accord with some preconceived world view. All observation is potentially contaminated, whether by our theories or our worldview or our past experiences.

Scientists, like anyone else, may be swayed by some preconceptions to look for certain experimental results rather than others. Scientists are people and suffer the flaws of humanity too, and it cannot truthfully be said that every scientific theory has arisen from a perfectly executed process of 'scientific method'.

Scientists have desires, opinions, and biases that may sometimes influence them in the selection of their data and hypotheses - even, very ocasionally, to the point of fraudulence. We trust that this is very rare, and that science is a self-correcting process with checks and balances - such as empirical replication, and peer reviews of published work. The scientific community as a whole, however, judges the work of its members by the objectivity and rigor with which that work has been conducted.

An infamous example of an experiment that couldn't be repeated to give the same results again, involved "cold fusion" - a supposed process by which nuclear fusion could be achieved at room temperatures instead of the extremely high temperatures normally needed for fusion to occur. The researchers involved in the original experiment were found to have been less than objective in their methods.

            Scientific Method

Science is a method that allows one to obtain, with the highest degree of certainty possible, reliable knowledge about nature. Scientific method establishes the principles that guide scientific research and experimentation, Scientific method is practiced within a context of scientific thinking, which is based on three things:

a)  using empirical evidence (empiricism),
b)  practicing logical reasoning (rationalism),
c)  and possessing a skeptical attitude (skepticism).

Very generally, a scientist begins with some observations (data); through some creative process a hypothesis is generated about how these observations came to be (e.g. a conceptual or mathematical model). This hypothesis is used to predict something that was not part of the original data. An experiment devised and executed to see if the predictions of our hypothesis hold true. If not then we are faced with several possibilities, e.g. our hypothesis needs to be revised, the experiment was carried out incorrectly, or the analysis of the results from that experiment was in error. 

Observation :- Observe and describe some phenomenon or group of phenomena.

Hypothesis :- Formulation of a tentative description, called a hypothesis, that is consistent with what you have observed. The hypothesis often takes the form of a conceptual model, causal mechanism or a mathematical relation.

Prediction :- Use of the hypothesis to predict the existence of other phenomena, or to predict quantitatively the results of new observations.

Experiment :-Performance of experimental tests of the predictions by several independent experimenters and properly performed experiments.  

Analysis :- Comparison of the experimental results with theoretical predictions, and attempt to reconcile differences, e.g. by modifying the hypothesis or repeating the experiment under different conditions