Retrospective Survey:

In a retrospective survey, the researchers may  also like to investigate possible cause and effect relationships by  observing some existing consequences and  looking  through the data  from  the past  for plausible causal factors. This  is different from experimental approach where data  is collected in  a controlled situation. For  example, you wish  to identify the  factors characterizing persons having  either high or  low accident rates. You  need to collect data of all the events of  interest (from the past as well  as present) occurring in certain subjects having high and  ' low accident rates. You would  then  examine the data by  going back, finding out  relationships and their meaning.

This is also called as  causal comparative approach, which  is "Ex-post  facto",  innature i.e.  data are collected by  looking to past events. In social behaviour studies, it is difficult to find a  cause for an effect because  there are many variables or factors that influence behaviour or occurrence of events.