136 PSYCHOLOGICAL METHODS OP TESTING INTELLIGENCE 



ing to this field. Burt deals with but a small number 

 of cases: one group of 30 l elementary' school- 

 pupils and another of 13 ' secondary ' pupils ; Eies has 

 investigated five classes in an elementary school. 

 The very much more extensive and precise investiga- 

 tions of the Breslau teacher, Hylla, have unfortu- 

 nately not yet been completed. 



Burt (72) tested his classes with 12 different tests. 

 The rank-orders obtained for the different tests show 

 quite different correlations with the estimated rank- 

 order six tests over 0.50, six tests under 0.50. The 

 tests that show the higher correlations are mostly 

 those that pertain to attention, motor skill and mem- 

 ory. These tests and their correlations are shown in 

 Table XVIII. On the other hand, the tests of dis- 

 criminative sensitivity uniformly show very low cor- 

 relations with intelligence a result worthy of note 

 because there still prevails a tendency in many quar- 

 ters to use sensory tests for testing intelligence. 



TABLE XVIII 

 BUST'S EXPERIMENTS WITH NORMAL CHILDREN 



/Correlation with-^ 



Est. Intell. 

 Elem. Secndry 



Test School School 



1. Dotting. (A zig-zag row of dots traveling 



at constant speed must be hit with a 



pencil) 0.60 0.84 



2. Spot pattern. (A group of dots to be re- 



produced by drawing after 5 exposures 



in a tachistoscope) 0.76 0.75 



3. Mirror. (A pattern visible only in a mir- 



ror is to be pierced at marked points) . 0.67 0.54 



4. Memory span for concrete and abstract 



words and nonsense syllables 0.57 0.78 



5. Alphabet. (Cards with the letters of the 



alphabet are to be properly arranged) . 0.61 0.80 



6. Sorting (50 playing cards of 5 different 



colors are to be sorted into 5 packs) . . . 0.52 0.56 



Resulting rank-order for all 6 tests. . 0.85 0.91 



