My impression is that data does not encompass the scientific models which are fitted from them, or which they are compared against.

Is there a word for the superset of data/datasets/measurements and models (ie. statistical models like a regression model or parametrizations of a physical phenomena)?

A possible example:

The phenomenon is poorly described by your [model and/or data input to the model].

  • 2
    I doubt it. Data is absolute. A model is basically an abstraction of this: an approximation. They are intrinsically linked but entirely separate things.
    – Michael
    Mar 8, 2016 at 12:22
  • 1
    I think the proper term is "data and model". Or perhaps "model and data".
    – Hot Licks
    Mar 8, 2016 at 13:02
  • I'm not sure I understand you. A phenomenon is a pattern you identify in the data. Say, global warming is a pattern in temperature records. So what do you mean with "the phenomenon is poorly described by the data?" Do you mean you believe the phenomenon is real, but you don't have the right data to support your belief?
    – Jacinto
    Mar 8, 2016 at 15:04
  • @Jacinto Fair point. There are however phenomenon you would expect by applying currently accepted theories in science, for example the quark-gluon plasma or higgs boson of the standard model of particle physics, or black holes and gravitiational holese of the general theory of relativity. These phenomena which we expect, may not behave exactly as we expect them from models, and/or the data which one tries to explain them with may not be adequate, mostly due to lack of accurate/bias-free data.
    – M.T
    Mar 9, 2016 at 8:19

1 Answer 1


From the observation you collect datas which then allow you to try to build a model. If this one is reliable, any new data shall comply within the scope or range of your model.

The word for such supersets is system ; a system depends on a theory.

  • I think this is a good answer if one has both a model and its data, but it does not seem intuitive to me that you could use this word about data or models by themselves. Do you have any thoughts on this?
    – M.T
    Mar 8, 2016 at 12:47
  • Model or system are correlated but data(s) lack the correlation rule that makes an order and organized set, because mere data is event.
    – DAVE
    Mar 8, 2016 at 12:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.