18 
Anyone who would analyse the figures from which a particular mean 
has been calculated, should turn, in the first place, to the appropriate 
Frequency Distribution Table, where he will find the standard deviation 
(o'), the probable error of the mean (E^), the coefficient of variation (V), 
and the number of cases examined (N), all duly recorded; and, subsequently 
— if he would pursue his investigation further — let him turn to the appendix, 
where the particulars of each individual on which this report is based, are 
fully set out. 
The similarity between some of the measurements and the contrast 
between others are self-apparent and call for but few comments, because 
table II displays these similarities and contrasts concisely and in tabular 
form. 
It may, perhaps, be of help to some in the understanding of this 
report, if an explanation is offered as to how the table of means and prob- 
able errors of means is to be interpreted, for, unless the significance of the 
probable error be appreciated, the figures in this, and in certain subsequent 
tables will fail to convey their meaning. 
In this table, No. II, it is stated that the mean or average stature of 
the men at Island lake is 170*0 cm., and of those at Oxford House, 172*5 
cm. The Oxford House men are, therefore, on an average 2 *5 cm., that 
is, one inch taller than the Island Lake men. But what reliance may be 
placed on these figures? At Island lake sixty-eight and at Oxford House 
fifty-five men were measured. These numbers are fairly substantial; the 
measurements were carefully taken; they were checked as they were 
taken; the men were not selected, but were picked entirely at random. 
With the expenditure of much time and labour it has been ascertained 
that the probable error of the mean stature of the Island Lake men is 
±0*48, and of the Oxford House men ±0*65. How probable errors are 
calculated need not concern us; we may accept them and proceed in the 
following manner to employ them : 
Stature and probable error Oxford House men 172-5 + 0 -65 
Island Lake men 170-0 + 0-48 
2-5 
The difference in the stature is obviously 2*5 cm. If the probable errors 
0*65 and 0*48 be squared, the results are 0*4225 and 0*2304. These when 
added together become 0*6529, the square root of which is 0*8080. 
If this result (0*8080), which is known as the “probable error of the 
difference” of these two statures (P.E. diff.) be divided into the difference 
between the statures (2-5 cm.), the answer is 3*1. That is to say, the 
difference between the statures is 3*1 times greater than the probable 
error of that difference; or in other words, the ratio between them is as 
3 * 1 is to 1 * 0. On consulting an appropriate table of odds {See next page) 
it will be seen that this ratio, if translated into terms of odds, will read : 
“The odds or chances are 26 to 1 in favour of our finding a difference of 
at least 2*5 cm. (one inch) in the mean statures of the Island Lake and 
Oxford House men sustained had we been able to measure infinitely larger 
numbers than circumstances permitted.” Or it might be expressed thus: 
