Researchers use different methods to ascertain whether a given effect could be the cause of cancer:

• epidemiology (comprehensive, statistical demographic studies)
• testing on cells and animals
• exposure assessment (studies of the strength of fields to which different people have been exposed for a given period of time).

Epidemiology
Researchers can investigate whether large groups of people having lived close to power lines for a long period of time statistically get cancer more often. This is called epidemiology.

Epidemiology can point to a possible link but epidemiology studies alone are seldom enough. It is very difficult to work with data that could date back 25 years. There may be a selection bias, and it can be difficult to find out whether the participants have been subject to other exposures. If a statistical link is found, the cause could be another effect which the persons examined have in common. In order for statistical studies to yield fairly accurate results, there must be sufficient participants. This could prove difficult especially with diseases afflicting a fairly limited number of people and effects that are not very common.

Testing on cells and animals
Both types of testing are useful in proving whether an epidemiological result indicates that a certain effect is the cause of a disease. They are also useful if researchers suspect that a specific mechanism triggers a given disease. This allows researchers to perform testing involving, for example, magnetic fields under controlled conditions and rule out to the extent possible other disturbing effects.

One test is never enough, however. Science demands that the tests can be repeated with the same result in different laboratories. It can also be very difficult to assess whether results gained from animal testing apply to humans as well.

Exposure assessment
Measuring magnetic fields in a given place at a given time is not difficult. But we move around a lot during the day, and the fields keep changing. This problem can be solved by means of a dosimeter – a portable measuring device. However, when performing epidemiological studies going way back in time, researchers are forced to assess or calculate the strength of the fields to which the participants have been exposed in the time when they were subject to the particular exposure. Such studies can go 20 years back in time, and the assessment may therefore be subject to a great deal of uncertainty.