Talk:Dependent variable

If $f(x)=y$ then y is the dependent variable. -[unsigned by 204.50.150.168 at 11:43, 22 October 2004]

Regarding the first sentence: "In experimental design, a dependent variable is a variable dependent on another variable (called the independent variable)". I think trying to define one item by using a word of the item in question makes it unclear. E.g. "a dependent variable is a variable dependent on another variable..." It might be best to use another word other than dependent to make it a bit more understandable. Just my $0.02. -therearenospoons 02:38, 22 February 2006 (UTC)[reply]

The definition is obviously tautological. Two other definitions were used in earlier versions of this article, and neither of them had this problem. Why not return to one of them? They were:

a variable whose values in different treatment conditions are compared
a variable which is not manipulated by an experimenter

John FitzGerald 19:40, 20 March 2006 (UTC)[reply]

Response variable is the more common term. Dependent has fallen into disuse, because dependent and independent already have assigned meanings in statistics.--Theblackgecko 22:44, 11 April 2006 (UTC)[reply]

A Google search for combinations of dependent variablee and experimental design produced 285,000 hits, while a similar search with response variable sunstituted for dependent variable produced only 71,000, so dependent variable doesn't seem to have fallen into disuse. If you could qualify the assertion (for example, among whom has it fallen into disuse?) and add the other meaning of dependent variable to the article that would be helpful. John FitzGerald 03:00, 21 July 2006 (UTC)[reply]

Looks like there is some confusion from the overlap of the usage of dependent variables by statistics and in experiments. Would it be better to split this article in two? Littlepinkpig 00:37, 22 September 2006 (UTC)[reply]

Saving this text because it is useful, but not immediately relevant to the article. Perhaps it could be a "see also" item, or integrated into an existing page. LightYear 05:04, 22 November 2006 (UTC)[reply]

There are also continuous variables, for example if you were timing the beetroot experiment, time would be a continuous variable because it is never going to change, regardless of other variables.

In the first paragraph:

In analysis, researchers usually want to explain why the dependent variable has a given value

Perhaps it would make sense to change this to:

In analysis, researchers usually want to explain why the dependent variable has an observed value

Thoughts? Lbthrice 03:16, 28 January 2007 (UTC)[reply]

Good suggestion, but perhaps a certain value is better, since in analysis, as apposed to physical science, there may be no guarantee that the dependent variable has been observed as such. Perhaps it was calculated, suggested, reported or otherwise came about. --LightYear 01:00, 29 January 2007 (UTC)[reply]

cleanup[edit]

See talk:independent variable. Michael Hardy 22:45, 9 March 2007 (UTC)[reply]