我也有类似的情况，有一个变量的系数大于1，计算出的边际效应也大于1了，什么原因啊

显著性检验的结果与自变量的单位无关。

http://www.stata.com/support/faqs/stat/mfx_size.html

I am using a probit model, and margins says that my marginal effect is greater than 1. Can that be correct?
Title    Marginal effects of probabilities greater than 1  
Authors  May Boggess, StataCorp
Kristin MacDonald, StataCorp  
Date  April 2004; updated July 2009