22

II

A  FIRST  EXPERIMENT
The  general  purpose  of  this  experiment  was  to  discover
a  set  of  mental  tests  which  could  be  used  by  the  employment ­
  office  in  selecting  applicants  for  certain  kinds  of
work.  Just  where  and  how,  specifically,  this  was  to  be
done  was  a  somewhat  hazy  problem  at  the  outset.  Hitherto, ­
  the  work  of  the  experimenter  had  been  confined  to
the  orthodox  experiments  of  the  psychological  laboratory.
Factory  conditions  and  factory  problems  were  therefore
novel  to  him,  no  more,  however,  than  his  purposes  and
methods  were  to  the  factory.  In  the  midst  of  this  somewhat ­
  hazy  and  intricate  problem,  four  distinct  conditions
were  discernible.  First,  in  finding  the  proposed  set  of
tests,  it  was  primarily  necessary  to  go  out  among  the
shops  and  make  a  general  survey  of  the  types  of  work
for  which  applicants  were  being  chosen.  Secondly,  it  was
necessary  to  make  an  intensive  study  of  one  or  two  operations ­
  at  the  start,  rather  than  an  extensive  study  of
a  large  variety  of  operations.  Thirdly,  a  large  number
of  workers  all  engaged  in  the  same  work  had  to  be
studied  in  order  to  make  as  wide  a  series  of  observations
as  possible  (obviously,  the  tests  tried  on  a  hundred
workers  would  give  more  conclusive  results  than  those
tried  on  only  ten  or  twenty).  Fourthly,  it  was  advisable
to  try  out  these  tests  where  the  work  was  simplest
and  most  automatic,  on  the  assumption  that  the  more
standardized  the  work  the  more  easy  it  would  be  to  discover ­
  a  standard  set  of  tests.