Pearson refuses to correlate

Continuing to read the 1905 Literary Digest…

Karl Pearson, the master of stats, has a strangely familiar complaint.

“At least 50 per cent of the observations made and the data collected are worthless, and no man, however able, could deduce any result from them at all. In engineer’s language we need to scrap about 50 per cent of the products of nineteenth-century science.” Pearson specifies as notoriously inaccurate meteorologic and medical statistics. Biological and sociological observations are ordinarily even of lower value. He doubts whether even a small proportion of the biometric data being accumulated in Europe and America could by any amount of ingenuity be made to provide valuable results, and believes that the man capable of making it yield them would be better employed in collecting and reducing his own material.

120 years later, no improvement. The same SPECIFIC PROBLEM is still present and often discussed. We like to blame the web and RUSSIAN_HACKERS and JANUARY_6_MALINFORMERS, but those things didn’t exist in 1905.

Why do we keep gathering and sorting shit? Damn good question. We’d be better employed in collecting and reducing our own personal observations, even if “subjective”.

Science should be a hobby, not a justification for genocide and torture.

%d bloggers like this: