In article , Cecil Moore
writes
Prometheus wrote:
Of course a proportional function is more likely although perhaps not
linear however your statements that "No tumors were associated with
less than 10 years of cellphone use" and "People with more than 10
years of cellphone use suffered twice as many tumors as non-cellphone
users" describes a step function at ten years {t = 1 for y 10 and t
= 2 for y 10, where y is the number of years and t the base rate for
brain tumours in a non-mobile phone using control group). Given this
simplistic step function I must have doubts about the accuracy of the
study that lead to such a conclusion

Your basic ignorance of statistical data is showing. What if, starting
at ten years of use, 1% of cell phone users suffered 1% more tumors
than non-cellphone users and a year later, 2% of cellphone users
suffered 2% more tumors than non-cellphone users, and a year later 3%
of cellphone users suffered 3% more tumors than non-cellphone users
... That is certainly ***NOT*** a step function, to which you objected,
but a ramp function that is certainly something to be concerned about.
Hint: I'm surprised that you don't know that nothing changes
instantaneously in reality, i.e. a 'step function' is purely an
invention of the human mind.

If, as you state, it went from nothing below ten years to two times
after then there was a step, it is that I object to; perhaps you do not
understand that it is a step {t = 1 for y 10 and t = 2 for y 10,
where y is the number of years and t the base rate for brain tumours in
a non-mobile phone using control group)

Hint: YOU quoted a single value for less than ten years and single value
for above, maybe you do not understand that is a step, do I have to draw
a graph of your statement.

t

2 ----------------------

1 ---------------------------
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 y

Can you see what happened? Of course I object to the function you quoted
as being improbable.

Your example of a proportional relationship is not justified from your
quotes, is irrelevant, and being a deception has no place in a
discussion of this nature.
--
Ian G8ILZ