Here we explain the steps involved in the scientific method

Observation or Idea

An observation of a phenomenon by a scientist is always the first step in the scientific process.   Commonly the observation may involve some aspect of nature or could be based on anecdotal reports.  However the observation is made, it is important that it is properly defined.

Example:  What causes leaves to change colour?  Is it due primarily to the change in temperature, a change in sunlight, change in rain fall or some other phenomenon?

Hypothesis

From the observation of a phenomenon, a hypothesis is formed.  The hypothesis is best described as a tentative description or a working assumption.  The hypothesis is then used to make predictions of probable effects.

Example continued:  We will form the hypothesis that leaves primarily change colour due to the change in sunlight.  Having made the hypothesis or assumption, we now need to research trees.  One needs to understand the mechanisms involved and how best to test them.

Experiments to Test Hypothesis

Experimentation is the most important step of the process.  Much effort is made  by the scientist at this stage to properly design or “control” an experiment so that it could either prove or disprove the hypothesis. 

The experiment may test the hypothesis directly or may test for consequences derived from the hypothesis using mathematics and logic.  As in every experimental science, “experiment reigns supreme”; experimental verification or proof of a hypothetical prediction or observation is absolutely necessary. 

If the “controlled” experiments do not bear out the hypothesis, it must be rejected or modified.  If the experiment is poorly controlled, the results cannot be used as support for the hypothesis.

Example continued:  The experimentation stage is always the most difficult.  One needs to design the experiment such that each agent that affect the growth of the tree is carefully controlled (water, sunlight, temperature, etc).

Peer-Review

The peer review process is a series of challenges and justification between the reviewers and the researcher.  Peer review is designed to ensure that publications meet acceptable scientific standards and filter out unsupported interpreatations and personal opinions. All aspects of the research are scrutinized at the peer review stage, for example the controls are critically re-examined and evaluated as to whether they were sufficient to support the results. Sense About Science - Guide to Peer Review Process

Research deemed to be properly conducted usually goes on to be published in a peer-reviewed scientific journal.  Publication in a peer-reviewed journal is a major accomplishment for researchers.  Poorly conducted research or experiments that appear to have been overly influenced by outside factors are given little scientific weight and are usually not publishable.  Poorly conducted research is rarely considered for inclusion in health risk assessments.

It is important to note that not all peer-reviewed journals use the same rigorous standards as the best journals such as Science and Nature.   Some research that is published in a peer-review journal may subsequently by rejected for inclusion in health risk assessment upon closer scrutiny or when other information invalidates the conclusions.

Replication

Once published, the study will undergo further intense scrutiny by a wider group of scientists who will go on to either repeat the results, refine the results, or disprove them outright through replication of the experiment.  Replications may try to reproduce the original experiment as accurately as possible or modify certain conditions to better control for factors that might influence the results. In the latter case, it may be described as a confirmation study.

Independent replication is essential to minimizing the effects of personal and experimental bias.  If the results are consistently replicated, then, the hypothesis becomes a theory and much greater weight is given to the original results.  If the results are inconsistent with the original, the weight of evidence reduces accordingly.

The advantage of replication is that it is unprejudiced, one does not need to believe the researcher or the results. One can simply redo the experiment and determine whether the results hold true. 

Note:  Exact replication is not necessary, but the results need to be confirmed by independent labs or be in agreement with other similar studies.

Additional Reference - Weighing the Evidence in EMF Health Research

Whole Body of Scientific Evidence Reviewed by Experts – The Weight of Evidence

Health Risk Assessment & Exposure Standards

As research is published, it is incorporated into health risk assessments.  The evidence provided by each study is carefully considered based upon its limitations and plausibility.  The exposure standards designed to protect public and occupational health are derived directly from the conclusions of this process using a prescribed level of precaution.

Scientific Expertise, an example:

With respect to handset and base station exposure, the International Commission on Non-Ionizing Radiation Protection (ICNIRP) evaluates regularly all relevant research and would revise the exposure guidelines if necessary.   The review comes from careful consideration of all available peer-reviewed research including clinical, epidemiological, animal, cellular, mechanistic, dosimetric and other studies.  The review conclusions will seek to sum up the consistency and accuracy of results and mechanistic plausibility of the reported effects.

Additional resources... 

Additional Information on Science

Health Authority & Educational Resources

• EMF Research - World Health Organization
• Promoting Good Science & Evidence for the Public - Sense About Science
• Guide to Science that makes the news - National Health Service (UK)

Other Resources

• Scientific Process Explained - GSMA
• Health Research - GSMA
• Good Science - Weighing the Evidence in EMF Health Research - GSMA & MMF 
• EMF Research - Mobile Manufacturers Forum