TYPE | error refers to :
a) The researcher found a significant relationship when none existed
b) The researcher found no significant relationship when none existed.
c) The researcher forgot to include the variation variable.
d) The researcher found no significant relationship when one existed.