TRANSACTIONS OF SECTION L. 809 



mill, which, in consequence, expresses a psychologically meaningless result. This 

 is especially apt to occur in the case of correlations, in the calculation of which 

 different observers so frequently disagree. 



A subsidiary cause of this discrepancy is due doubtless to racial differences 

 in correlation. The racial differences which undoubtedly exist in the correlations 

 of physical measurements almost certainly have their counterpart in respect of 

 mental measurements. So the correlation in a heterogeneous people may in one 

 sample of it be high, in another be low or reversed ; in one sample the probable 

 error may be great, in another small, according to accidental differences in racial 

 diversity. 



But the main cause lies in the neglect of the introspective element. The only 

 way to ascertain what is being tested by psychological experiment is to have 

 recourse to the subject's experience. Owing to neglect of this a test of mental 

 fatigue may not involve mental fatigue at all; it may in different subjects 

 involve the play of complicating factors (automatism, boredom, duty, ambition] 

 in different degrees. So, too, a positive correlation between general intelligence 

 and sensory discrimination may be due to the fact that the very nature of the 

 test has compelled the subject to use his intelligence while carrying out sensory 

 discrimination. To avoid spurious measurements and correlations of this kind, 

 too much care cannot be taken to find out exactly what factors the experiment 

 involves ; and this can only be done by individual introspection, which is 

 impossible in the blind wholesale collection of data by untrained observers. 



We are prone to hope that statistics will yield results out of all proportion 

 to the value of the material put into them. Only collect enough data and a valu- 

 able conclusion will emerge ! Nothing can be farther from the truth, especially in 

 regard to ' mental tests.' The treatment of such data en masse is apt to give 

 a mere blur, which hides mental processes actually existent and claims to reveal 

 others that have no existence. 



Mass-experiments, however, have their use. In everyday life we do not care 

 how an individual works, how he knows; we want to know how much he can 

 work, how much he knows. For this purpose we require standards of pro- 

 ductiveness, standards of knowledge, which will differentiate, for example, the 

 feeble from the normal and will mark the progress of the former. But let us 

 clearly recognise that these are not psychological tests. Let us not spoil two good 

 things by pretending that tests of production are a measure of definite mental 

 processes, for from the psychological aspect the results are a mere blur. 



MONDAY, SEPTEMBER 5. 



The following Papers were read : — 



1. The Teaching of Handicraft and Elementary Science in Elementary 

 Schools as a Preparation for Technical Training. By J. G. Legge. 



The elementary school's right aim — a general education to form the basis of 

 specialisation later. So, too, with the secondary school, though there now recog- 

 nised that there is no one type of general education, and consequently both at 

 home and abroad various types, each with a particular colour or bias, are being 

 developed. Why not similar variety of types in the case of elementary schools ? 



The particular variety under discussion to-day — a practical curriculum for boys 

 in a town school, leading naturally on to technical education. Part played by 

 British Association, notably in Reports of 1906 and 1908, in establishing claim 

 for introduction of handicraft and experimental science into the curriculum. 

 Impetus towards this given by Board of Education's recent Memorandum on 

 Manual Instruction in Elementary Schools. 



The scientific basis for the practice of manipulative work — development of 

 brain centres presiding over co-ordinated movements of all kinds. Variety of 

 purposes in view when principle thus established is advocated, and consequent 

 variety of means for applying it. No one single purpose in view, and therefore 

 no one single method of applying principle possible. 



