THE INCREASE OF IGNORANCE 



Wards of Pittsburgh with Most Illiterates and Most Foreign-Born Have High 



Birth-Rate and Also Show Lower Infant Mortality than Some 



of the Best Educated and Prosperous Wards 



Till'; Im)IT()k 



IT is the function of Art to isolate 

 different aspects of life and hold 

 them up for observation. The on- 

 looker must not "j^o behind the 

 returns." If you are lookinj^' at the 

 Wnus de Milo in an esthetic frame of 

 mind, >-ou must see nothinjj; but a 

 statue and an idea in it. If you begin 

 to wonder about her pedigree or whether 

 the Greeks really had such straight 

 noses as she has, you arc no longer 

 studying art, but science. For science 

 deals not with things as they are, but 

 with things in relation to each other. 



One of the most ingenious and 

 satisfactory ways of showing the rela- 

 tions of things to each other is by 

 a])ijlying the calculus of correlation, a 

 branch of higher mathematics which 

 has been extensively developed during 

 the last generation. Isolated statistical 

 facts which seem to have little or no 

 meaning can be treated with it and at 

 once take on life — at once allow the 

 student to project the picture forward, 

 so that he sees not onh' what is now 

 ha]j]jening but what wnll ha])ix'n in the 

 future, under certain conditions. The 

 accuracy of the ]jicture is limited only 

 by the exactness of the original facts. 

 The i)resent imjx'r undertakes to give a 

 rough illustration of how the calculus 

 of correlation may be a]:)]:)lied to the 

 vital statistics of a city to bring out 

 their significance. The city chosen is 

 Pittsburgh, from which Prof. Roswell 

 I lill Johnson has turned over to me some 

 data which Miss Sadie Scorer, one of 

 his students, comi)iled a few years ago, 

 but did not use. They show the l)irth- 

 ratc, infant mortality, number of illiter- 

 ates, j^roporlion of native whites of 

 native parentage, and various other 

 facts aV)out eac-h f)f the twenty-seven 

 wards into which (ircnlcr Pittsburgh 

 is divided. 



178 



I have never been in Pitlsljurgh. but 

 as a newsi3a])er reader I picture it as a 

 huge, smoky, industrial center with 

 something like three-quarters of a 

 million inhabitants, and containing great 

 extremes of wealth and ]jovert\', innu- 

 merable millionaires on their magnificent 

 estates, and vast, overcrowded c}uarters 

 where tall chimneys of factories jostle 

 each other, and where a population of 

 laboring people, largely foreign-born and 

 ignorant, lives on the edge of the 

 financial dead-line. 



An inspection of Miss Scorer's figures 

 as they stand is sufficient to give me 

 some idea of the make-up of the metrop- 

 olis. But I want to watch its vital 

 ] jrocesses ; to sec the changes taking i)lacc 

 year by year and thereby get an idea 

 of what it will be in the future; just as 

 the surgeon studies the case history of 

 his ])atient and arrives at a prognosis. 

 I"or this inir])ose nothing is more con- 

 venient than the calculus of correlation. 



CORREL.\TIOX 



It is a matter of common observation 

 that there are all sort of correlations in 

 daily life ; that certain things go together, 

 and that an increase in one factor will 

 necessarily cause an increase in some 

 other. For exami^lc, if the po])U- 

 lation of a city increases, the 

 amount of food consumed in the city 

 will also increase; the two facts are 

 closely correlated. And one might also 

 guess that as the number of births 

 increases, the number of deaths ^vill 

 also increase; there are more people to 

 die, and die they must, soon or late. 



Now the exact amount of association 

 between two things is nicely measured 

 bv the calculus of correlation, and 

 expressed in a coefficient which is a 

 (Iceimal fraction somewhere between 

 and -|-1 or —1. If the corrclrition is 



