─────────────────┬─────────────────┬─────────────────┬───────────────── │ Non-Deficient │ Deficient │ Total ─────────────────┼─────────────────┼─────────────────┼───────────────── Non-Delinquent │ 22,305│ 109│ 22,414 ─────────────────┼─────────────────┼─────────────────┼───────────────── Delinquent │ 268│ 4│ 272 ─────────────────┼─────────────────┼─────────────────┼───────────────── Total │ 22,573│ 113│ 22,686 ─────────────────┴─────────────────┴─────────────────┴───────────────── The total number of boys is taken from the census of school children for 1915-16 compiled by the attendance department of the Board of Education. It includes those in public, parochial and private schools and those not attending. The number of delinquent boys is taken from the report of the Juvenile Court of Hennepin County, Tables H and I. The number of repeaters and the proportion of delinquent cases dismissed at the hearing are subtracted from the total number of new cases. The difference between a correlation of .29, the highest I found, and .59, Goring's lowest result, indicates that conviction for felony in Great Britain is more closely associated with deficiency than juvenile delinquency is associated with deficiency in such communities as Minneapolis. It is to be remembered, however, that Goring's calculation gave the convicts a life-time in which to be convicted, while ours gave the boys only 16 years. The relation of potential delinquency after 16 years of age to deficiency might be greater among Minneapolis males than the corresponding relation we found among the boys; but the difference in these correlations is more easily explained by supposing that the type of serious delinquency represented by sentences to penal servitude, in England at least, is more closely related to deficiency than are the lighter forms of delinquency found among the youth of an American city. The most significant fact demonstrated by the correlations between juvenile delinquency and deficiency is that there is a positive relationship which is significant in amount. With the maximum estimate the correlation is nearly 6 times its error. This is the first time that the relationship has actually been calculated in connection with any group of juveniles. We can say that when a Minneapolis boy is below the average in tested ability for his age, he is most likely to be .16 to .29 of the same amount below the average in legal conduct, both measurements being in corresponding units.