out of school, at home, with playfellows, or at work we cannot be sure that other external influences might not ultimately be found to be more important in connection with juvenile delinquency than either the school life or mental incapacity. The further consideration of the causes of delinquency we shall now make the subject of a broader inquiry. ----- Footnote 32: Mental ages VI and VII regarded as satisfactory for the first grade, etc. CHAPTER XI. DEFICIENCY AS A CAUSE OF DELINQUENCY In a preceding chapter we have shown the frequency of tested deficiency among various types of delinquents. We may now further consider the significance of this association of delinquency with deficiency. The best plan for discovering its meaning is provided by the technical method of correlation. The data in the published reports of the score or more of investigations which I have reported is wholly inadequate for following out this method. We must, therefore, for the present content ourselves with noting what has been discovered by the better analysis of similar data which was supplemented by the necessary information as to the distribution of the different types of crime in the corresponding general populations. To this we can add certain correlations in connection with the small Minneapolis group of tested juvenile delinquents. We are indeed fortunate to have the fundamental work of Dr. Charles Goring on “The English Convict,” from which to formulate a point of view regarding the relation of deficiency and delinquency. This work represents ten years labor in making observations, collecting, tabulating, and statistically evaluating data on 3000 convicted men, who were found in the English convict prisons where they had been sent after conviction in the higher courts because guilty of grave or repeated offenses. It was carried out with the co-operation of a corps of workers who had the help of Professor Karl Pearson and his assistants at the Biometric Laboratory of the University of London, in the statistical reduction of the almost overwhelming mass of data. By the large use of partial correlation the _relative_ influence of various factors upon criminality was investigated as it never had been before. It is, of course, not possible to reproduce here the conclusions of this monumental work which should be made more widely available in the libraries of this country. We shall, however, select certain conclusions which bear most directly upon our problem and which rest upon well established statistical deductions, and compare them with a few other studies which have contributed interesting side lights upon the causes of delinquency. A. THE CHANCES OF THE MENTALLY DEFICIENT BECOMING DELINQUENT.