SECRET STATISTICS.

Dominion, Volume 5, Issue 1259, 16 October 1911, Page 3

SECRET STATISTICS.

Sir, —AVhen Mr. Gammcll commenced digging for statistics two years ago bis objcct «'as to unearth the statistics that it was alleged hail previously been discovered by Sir Robert Stout. Whenever a. few bones appeared as the result of his operations ho seemed to emerge with jubilant cries of "Eureka!" The sce.ret statistics were supposed to bo in every (shovelful thrown out. But oven liow ho has not obtained them, and his "Eurekas" now seem to be mingled with cries of. "All my work!" The statistics he digs out are now his own, and 1 am invited to dispute tiiem, and told that until I do <■<> ] have no right to demand the produit'on of Sir Hobert Stout's, which, < i course, are to remain in nnl.ibu. rv.-,i nfter Mr. Gammell's two .vmiV .Vibrate digging. I do not to i)"nv that Mr. Clammell ha.s Iri:t!<!u.'i.v 'ju.il-.-'l his figures from official Hui. ! :lo deny his conclusions. .V Hir l,\.hert's, the very cxistoniv of the -:-ii:-Mc-. he relied u|xmi are in quv-d't'., fo ny nothing of his conclusions. A-s-iPiir, •' they o'iet* did e\i.»t, they h.ivc 'louh! been si> completely oyriti;rr,r,i • 11., controversy en the subject tlot 'liin.-nris are now made for some absurd, inimmssajy, or irrelevant, thing to be done before they can bo brought out. Perhaps even the compiler of our l'ear Book has buried tlioni uowii too deep even for Mr. Gammell ever to reach again. I refe.rrcd Mr. GainmcU in ray last to liottlpy's Elementary Manual of Statistics. [[•.! will find there tho great gap between Itjs own figured and conclusions pointed out. Oji Pago W tho writer warns liia;

"It is novcr safe to take published statis- - tics at their face value, without knowing their meaning and limitation, and it is always necessary to criticise arguments that nro based on them, unless ono is able .to trust implicitly tho knowledge and good faith of the persons bringing them forward. Jt is extremely easy to falsify the lessons which numerical statements should teach. The actual use and appreciation of statistics is ultimately a matter of intelligence, special knowledge, and common-sense." Again, on Page 5: "Statistics only furnish a tool, necessary though imperfect, which is dangerous in the hands of those who do not know its nso and limitation." Secularism must indeed be in evil plight when its defenders are compelled to act as if their only arguments are based on statistics, and doubtful subjective conclusions drawn from statistics. Conclusions that, in fact, can be turned either waywith equal facility, according to the selections made. My object is nob a controversy with Mr. Gammell. but to draw attention to the fact that Sir Robert Stouts assertions, bused on statistics which even a two v-earsr 1 search by Mr, Gammell have not brought to light, and which evidently are in nubihus, are us«d now in England as a convincing argument for secularism.—l ara ' CtC " PHYLAX. October 3, 1911.