Sorry
That's Incorrect

No. Actually, an ex post facto procedure (after-the-fact)
is not really a research design, and it has major problems
that prevent any causal, correlational, or contingency
inferences. The problem is that only one variable is observed
or measured. No independent, differential, or correlational
variables are included,  no manipulations are made, and no
controls are applied. About all one can do if data have been
gathered this way is to examine the data to see if they have any
descriptive value at all of some current variable. Such data
are also of questionable reliability. Ex post facto procedures
do not provide much useful information.