MACHINE  OPERATORS

II 9

earnings  of  these  operators,  averaged  over  periods  of  from
°ne  to  four  months,  depending  on  the  shop  and  the  number
°f  months  of  production  available.  In  one  of  the  three
shops  a  night  shift  had  recently  started  work  and  tests
'Were  given  to  twenty-one  women  on  this  shift.  Many  of
these  women  were  immigrants  who  could  not  understand
English,  and  quite  a  number  were  elderly  women.  Howler, ­
  comparatively  little  difficulty  was  experienced  in
showing  them  how  to  perform  the  test.  The  results  in  all
oases  showed  that  there  was  a  marked  correspondence  between ­
  ability  in  this  test  and  ability  to  feed  dial  machines.
The  results  in  all  groups  may  be  summarized  as  follows:

Correlations

Number
of  operators
tested

1st  trial

2nd  trial

3rd  trial

^hst  Shop.  .  .

25

■36

.40

•54

Jwcond  Shop....

22

•43

.14

no  trial

Third  Shop

28

■4i

.28

•43

^bird  Shop.  .  .  .
_  (night  shift)

21

.22

.22

•50

The  correlations  found  are  not  high  in  every  case,  but
^  the  correlations  obtained  on  the  third  trial  are  uni-°tmly
  good.  This  may  be  due  to  the  fact  that  the  first
trials  were  too  easy,  whereas  the  third  trial,  with  the
°t  only  half  open,  more  nearly  approached  the  actual
°P e ration  in  difficulty.  It  also  seems  likely  that  some
°P e tators  were  so  set  in  their  regular  habits  of  machine
°P er ating  that  they  could  not  readily  adapt  themselves
^0  the  changed  conditions  of  this  test.  The  interference
r  P r evious  habits  is  a  well-established  psychological  prin ­