Re: some thoughts on science
"Dale Kelly" <dale.kelly@comcast,net > wrote in message
news:pan.2007.04.20.01.47.39@comcast,net ...
> The historical roots of the word science are related to the Latin term
> "Scientia", meaning "knowing". But this is not how science is used in
> practical purposes. Inference and philosophy are absolute knowing.
> Science in modern day purposes is a best practices way of dealing with
> conjecture. In the scientific process, a conjecture, is a hypothetical. A
> hypothetical is most often called a hypothesis. If a hypothesis is
> reproducible, then it is testable, and called a theory. If a theory has
> been tested, it is said to be founded.
Your explanation is so confused, it might be best to ignore it and go back
to the basics.
The scientific method is a four step, cyclic method.
Step 1 is data. Data is the result of observation and experimentation.
Step 2 is a theory. A theory is an explanation for the data.
Step 3 is a hypothesis. A hypothesis is a specific prediction based upon a
theory. (If you have difficulty with this use of "hypothesis", just call
Step 3 a specific prediction based upon a theory.)
Step 4 is an experiment. The experiment directly tests the hypothesis. The
result is more data and we return to Step 1.
The validity of any theory is determined entirely by how well it fits the
data. If new data is produced that contradicts a theory, then the theory
must either be altered or rejected. Experiments in science should, in
principle, be repeatable. The results of experiments should be independent
of the experimenter.
> If a theory has not been tested,
> that theory is said to be unfounded. Statistics is the language of
> science. An analysis of variance will tell you, if you have designed your
> testing properly, the confidence you can have in the data you observe.
> There are two major considerations here. One is that if you have not
> included all variables in your testing, the variance of your data will
> result in a low confidence for your results. This means to get a high
> confidence interval, some degree of solid inference has to be built into
> your hypothesis. So you see that nothing can really be known empirically,
> by testing, that is not first known in inference. The second
> consideration to the scientific process lies in exactly how the analysis
> of variance is carried out. A true analysis of variance, must include the
> variance of the measurement apparatus or observation equipment. And must
> also include the variance of how the observation equipment is, and the
> observation variance for that etc., therefore there can be absolutely no
> confidence in empirical data.
Analysis of variance may be a useful tool, in some cases, but it certainly
isn't a universal requirement of science. In fact, its applicability seems
to be limited. Perhaps you could give us an example of how analysis of
variance might be applied in some field of science (The more closely related
to evolution, the better).