Why there is never absolutely accordance between good theory and experiment result?

[ndung200790]

Please teach me this:
Despite perfect theories(e.g quantum theory) we never have a absolutely fitted between theoretical and experiment results.Why is it?So it seem that we are hopeless to wait a ''absolutely'' perfect theory?
Thank you in advance.

 

[ndung200790]

And we must to smooth the experiment data to accord with the theory,then I think is it the theory were not perfect?

 

[lugita15]

Blame it on the experimentalists! :tongue:

Even if we had an absolutely perfect theory which predicted all physical phenomena with infinite precision, our experimental instruments would still presumably have only finite precision, and would be subject to all kinds of statistical and systematical errors, so you'd still have plenty of experiments.

 

[ndung200790]

But I would like to say about the difference between the theory and experiment that greater than error limit in experiment.But after we ''smooth'' the data it fits with the theory.

 

[StevieTNZ]

But after we ''smooth'' the data it fits with the theory.

Sounds like taking into account systematic and statistical errors. We don't change data (well I hope not), just so it appears we're validating a prediction.

 

[mfb]

But I would like to say about the difference between the theory and experiment that greater than error limit in experiment.But after we ''smooth'' the data it fits with the theory.

Can you give an example for this?

I don't know what you mean by "smooth". Experimental challenges often require some sort of fancy analysis stuff to get results, but you don't modify the data. You just care about experimental issues.

A difference between theory and experiment which is too large to be explained by experimental errors would have a massive impact on physics (think about the OPERA experiment).

 

[Many_S_Theory]

The goal of science is to change our data to fit our hypothesis... wait, that doesn't sound quite right... Is your issue simply that with experiments come experimental error and only after errors analysis and lines of best fit do we recover the desired agreement with theory? There's nothing sinister about averaging or random error.

 

[f95toli]

But I would like to say about the difference between the theory and experiment that greater than error limit in experiment.But after we ''smooth'' the data it fits with the theory.

This really depends on the experiment. There are plenty of experiments where the theory agrees extremely well even with the "raw" data.
However, when we model real experiments there are usually (but again not always) phenomena we can not model very well; either because it is simply too diffucult (and would e.g. require a lot of numerical simulations) or because we do not know the values for the relevant parameters.
Also, real measurement setups are always a compromise, even if you have good, well-calibrated instruments you still have to deal with things like noise and drift.

The trick is simply to try to design your experiments in such a way tha these known unknowns play as little roll as possible.