The scenario in which a scientist who is studying earthquakes includes the impact of wind when performing some tests of damages to structures is an example of failing to hold all other things constant.
In every research and experiment, there are variables that are held constant in order to see how another variables influence the result of the experiment.