42 



Indiana University Studies 



to grade should be approximately the same for both. The cases 

 of "lynx" and "vise" illustrate this very well. Only in the fourth 

 and sixth grades is there a difference in error of as much as 3 per 

 cent. A scale made of words of this type will doubtless be 

 accurate to a very high degree for children whose previous train- 

 ing has been similar to that of those from whom these records 

 were obtained. 



As a matter of fact, it is not always easy or even possible to 

 get words which behave so nearly alike as do "lynx" and "vise". 

 Thus "tender", which has the same average per cent as "lynx", 

 does not approximate so nearly the "lynx" value as does "vise" 

 in grades 3, 4, 5, 7, and 8. "Peritoneum" is evidently more 

 difficult than "chancel" for grades 3, 4, and 5, but less difficult 

 for grades 6, 7, and 8. Some of the words seem highly 

 erratic in behavior. Thus "corporal" was missed by 73 per cent 

 of the eighth grade, but by only 56 per cent of the seventh; 

 "code" was unknown to 76 per cent of the seventh and 70 per 

 cent of the sixth; "sextet" reaped 71 per cent error from the 

 sixth and only 66 per cent from the fifth; "artery" was known 

 to 42 per cent of the fourth but to only 30 per cent of the fifth; 

 and altho 66 per cent of the third knew "shin" to be a part of the 

 body, only 59 per cent of the fourth grade were aware of that 

 fact. 



In general, however, these apparent inversions of difficulty 

 are not the rule; in fact, they seem astonishingly rare when one 

 considers the myriad factors which operate to produce linguistic 

 abihty. A more common limitation on the use of words for 

 measuring purposes is the fact that they play out, as it were, at 

 one or the other end of the scale. Thus, "hammer" has 36 per 

 cent of error in the third grade, but produces none at all above 

 the sixth. At the opposite extreme is "octile" with 80 per cent 

 error for grade eight, but producing 98 per cent error in the 

 third, far too much for practical use in measuring a class. 



Indiana-Thorndike Visual Vocabulary Scale 



5. arms (6) brown (c) chair (h) cow (4) dog (4) dress (we) 

 green (c) hat (uw) robin (bi) shoes (we) walk (ft) 



10. bear (4) bed (h) bone (6) bonnet (ive) bookcase (h) 

 court (g) dance (ft) ear (6) eye (h) fight (w) fox (4) hop (ft) 

 lark (bi) overcoat (ive) ship (s) step (ft) stocking (ive) swallow (bi) 

 table (h) yellow (c) 



