变因是为了厘清哪些因素能使实验产生不同的结果而形成的概念。分为下列三者：

此外，有一种实验标题能够透露出操作变因及应变变因为何，如「酸对秃头的影响？」中，「酸」为操作变因、「秃头」则是应变变因。但并非适用于所有实验标题，如「熔化热实验」、「制作氧气实验」等'没有[问号]的名称则无法透露操作变因及应变变因为何'或是'根本没有变因 (亦指无实验组与对照组)。

In mathematical modelling and statistical modelling, there are dependent and independent variables. The models investigate how the former depend on the latter. The dependent variables represent the output or outcome whose variation is being studied. The independent variables represent inputs or causes, i.e. potential reasons for variation. Models test or explain the effects that the independent variables have on the dependent variables. Sometimes, independent variables may be included for other reasons, such as for their potential confounding effect, without a wish to test their effect directly.