ONE SHOT CASE - STUDY

Campbell and Stanley classified this design as “pre-experimental.”  No variable is manipulated.  The researchers simply find some group of subjects who have experienced event X and then measure them on some criterion variable.  The researcher then tries to related X to O.  My kids were in the public schools here when a terrible tornado ripped through the county just south of our house.  After the tornado left, psycho-researchers descended on the schools, conducting research to determine the effects of the tornado on the children’s mental health.  Of course, they had no pretest data on these children.  Without a comparison group, observations like this are of little value.  One might suppose that there is an implicit comparison group, such as that provided by “norms” on the measuring instrument, but how do we know whether or not our subjects already differed from the “norms” prior to experiencing the X?

Thanks