2

I have seem the terms top-down bottom-up approach,top-down bottom-up models,top-down bottom-up causation in many papers. I am bit confused.Are they really same or how can we define them ?

Improve this question
2
  • If you're seeing these phrases in essays about topics in a relatively specific domain of discourse (philosophy of science, I assume, per your tags), the authors do probably intend a lot of conceptual overlap/point-of-contact for using them. The prosocial dimension of science favors researchers who want to talk with, not past, each other (and provides for one of the indicators of pseudoscience: intellectually antisocial behavior, like aggressively cryptic writing modulo the target audience, seems to set off our "pseudoscience alarm").
    – Kristian Berry
    Jul 15 at 15:51
  • See e.g. here
    – Mauro ALLEGRANZA
    Jul 15 at 17:20

2 Answers 2

Reset to default
4

If you're talking about the philosophy of science, 'top-down' approaches start with theory and then tries to produce observations through experimentation that validate that theory. 'Bottom-up' approaches begin by collating observations in the real world and trying to construct a theory from them. The terms are more or less synonymous with 'theory-driven' and 'data-driven'. In the philosophy of science it's generally seen as a chicken-or-egg question, since both approaches are frequently used in practice. It's mainly an academic question about which approach has primacy.

Improve this answer
2
  • I think you are telling about deductive and inductive reasoning
    – quanity
    Jul 15 at 17:33
  • @quanity: inductive and deductive reasoning play into the issue, yes.
    – Ted Wrigley
    Jul 15 at 19:55
3

A bottom-up approach to understanding something is one in which the whole (the thing to be explained) is conceived solely as an effect. A top-down approach is one in which the whole is conceived as the cause.

Improve this answer
4

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .