A. At the Glen Lake Farm School for Boys, Hennepin County, Minnesota.

We are now in a position to evaluate the Binet examinations of delinquents. Let us first note our results for a group of 123 consecutive cases at the Hennepin County Detention Home.^[14] It is not a detention home in the sense of a place where children are held awaiting the disposition of their cases by the Juvenile Court. It is better described by its unofficial title, The Glen Lake Farm School for Boys. This county training school for delinquents is located on a splendid farm beside a small lake fourteen miles outside of Minneapolis. The boys are sent there by the juvenile court for a few months' training as an intermediate discipline between probation and sentence to the State School at Redwing.

The character of this group of 123 randomly selected delinquents is further indicated by the fact that 69 of them had already been brought into court two or more times, 54 were first offenders. Boys are sent to Glen Lake whenever the nature of their delinquency or the conditions at home, together with the personality of the boy, seem to the court to require this special training. A summary of the offenses for which the boys were brought into court does not, therefore, show the character of the boy as it is known to the court through the evidence and the efficient service of the probation officers. It shows, however, that the last offenses for which this group were being disciplined were as follows: Petit larceny 29, truancy 25, incorrigibility 25, burglary 9, grand larceny 6, disorderly conduct 4, malicious destruction of property 4, trespass 3, sweeping grain cars 3, breaking and entering 3, indecent conduct 2, miscellaneous offenses one each 8, total 123. Perhaps a more important indication of the character of the offenders in this group is that they represent about a quarter of the cases brought before the juvenile court during the period of this study, a little over a year. With the exception of a very few cases sent directly to the State Industrial School they may thus be regarded as typically the worst quarter of the delinquent boys under 17 years of age in Minneapolis.

The majority of boys were tested by myself after several year's experience with the clinic in mental development at the University of Minnesota and after examining many other delinquents. Some were tested by assistants from the university clinic, Mrs. Marie C. Nehls and Mr. Harold D. Kitson, who had been specially trained for this. Their detailed reports were carefully gone over and evaluated. The Binet 1908 series (136) was used, except that for tests above XII either tests XIII were used, or later these were supplemented by two other tests, which have been placed in the age XV group or adult groups, in the revisions of the Binet scale published by Goddard (110) or Kuhlmann (135). This variation was of small importance since a boy was regarded as of passable intellect if he scored X.8. We always gave the three tests of the XIII group and the boy was credited with age XIII if he passed two out of the original XIII year tests or four out of five tests given above XII. In accordance with our conservative position the rule of this 1908 scale for scoring was followed and the boy credited with the highest age for which he passed all but one test, plus one year for each five higher tests passed. This is the basis of the 1908 form of the scale as standardized by Goddard. Appendix II gives the detailed results for each boy with exact life-age and tenths of test-age on the scale, basal test-age with the tests, grade in school at the first of September when he was of this life-age and offense for which he was being disciplined. It also indicates which boys were repeaters. The results of this table are summarized in Tables VIII and IX. The life-ages at the last birthday are used rather than the nearest ages, since this accords with Goddard's standardization and with the common use of the term “age.” Moreover it seems to conform to the best practise and to be less likely to lead to mistakes. Table IX also shows the school position of each boy. Since a number of the older boys had left school, in order to tabulate their school positions in reference to their life-ages it was necessary to assume that they would have continued to progress normally from the position they held when they left. The Minnesota law requires attendance at school until sixteen years of age unless before that the child graduates from the eighth grade. In this group most of those sixteen years of age and a goodly number of those fifteen years old had left school, so that their school position had to be advanced a year in the table; a very few of the 16-year-olds had to be advanced two years in the table. In all cases the school position is given relative to the first of September when the boy was of the life-age given. Either ages six or seven are taken as satisfactory for the first grade, ages seven or eight for the second grade, and so on with the other grades.

An Arabic numeral after a Roman numeral indicates the number of cases, when more than one case occurs at any position in the table. Parentheses indicate cases testing presumable deficient or doubtful. S is a satisfactory school grade.

The summary of the Binet scale testing of this group according to the valuation which we have adopted, shows two clear cases of tested deficiency. One boy who was 13 years of age tested VIII and was the only case sent to the State School for Feeble-Minded from this group. The other was 16 years of age and tested IX. Besides the two presumable deficients, seven other boys were uncertain according to our interpretation, as judged by the Binet tests alone. One of them was 13 and tested IX, the others were 14, 15 and 16 and tested X. This would make a total of 7% possibly socially deficient, since they were all delinquent. This seems to be the largest estimation of deficiency which would be justified on the basis of these test results. To show, however, how important is the interpretation of the results obtained with Binet examinations when treated in gross, it need only be stated that a few years ago, when this study began, it was not uncommon to count all who were retarded three or more years and testing XII or under as feeble-minded. On that absurd basis, there would be 45 such cases (37%). As we have considered at length the reasons for not counting a person as even of doubtful intellect who tests XI or above or is less than three or four years retarded, we do not need to rehearse them here.

Using our conservative basis for interpreting the results of Binet examinations, let us now review the evidence of the proportion of delinquents which is intellectually deficient. We shall compare the available data on groups of tested delinquents which have not been subjectively selected, provided that the data permit of restatement on the basis of the borderlines we have adopted. The evidence of tested deficiency on over 9000 objectively selected delinquents has thus been assembled under approximately the same interpretation of the borderlines. This should help to make it clear how extensive the preparations must be for dealing with this problem of the defective delinquent and where the needs are most pressing. It should also enable us to discover when the estimates have been excessive. We shall confine ourselves to the reports of objective test examinations, so that the estimates do not depend upon the judgment of the examiner alone. A bibliography of these studies is given at the close of the book. How much more has been accomplished in this field in the United States than abroad is illustrated by the fact that repeated search has failed to discover any reports of Binet examinations on representative, randomly selected groups of delinquents in any foreign country. Binet examinations have been made of juvenile delinquents in Breslau (34) and in Frankfurt a. M., and in London (56); but only upon selected cases.

Those who wish to compare the results as to tested deficiency with the subjective opinions of various estimators should consult the reviews of this literature by Bronner (6) and by Gruhle (121). The effect of such a comparison is an increasing conviction that it affords dubious evidence of the relative amount of deficiency in different groups of delinquents. Without objective tests, there is no means of telling what amount of mental retardation the different experts would class as feeble-mindedness.

(a) Women and Girl Delinquents in State Institutions.

Women in state penitentiaries are a small group among delinquents in institutions. According to one study by Louise E. Ordahl and George Ordahl^[15] the frequency of tested deficiency is smaller among them than among women committed to reformatories, who in general commit less serious crimes. All except one of the 50 women prisoners enrolled were tested with the Kuhlmann 1911 revision of the Binet scale. About half were negro women. Only 6 (4 negroes) tested IX or below and were in our group of presumably deficient by the tests. Twenty others (13 negroes) tested one Binet age higher and were in the doubtful group.

If we consider the worst condition so far as intellectual deficiency is concerned, we find it in the reformatories and training schools for women. Dr. Weidensall applied the 1908 Binet scale to 200 consecutive women, 16 years to 30 years of age, as they were admitted to the New York Reformatory for Women at Bedford. Seventy-seven tested IX or under and were within our presumably deficient group. An additional 74 tested X and were in the uncertain group, although if we regard them all as deficient because of their persistent delinquency, we have a total of 75% (59). These results were duplicated by Dr. Fernald (16). She tested 100 other consecutive cases with the 1911 scale and found 41% tested below X, our presumably deficient group. She regards these as “feeble-minded with certainty.”

Dr. Katherine Bement Davis, the former superintendent at Bedford, estimated herself that among 647 prostitutes who were inmates there, 107 were “feeble-minded (distinctly so);” 26 “border-line neurotic;” 26 “weak-willed, no moral sense;” 11 “wild, truant, run-a-ways.” This makes a total of 26% of this group whom she apparently thought might possibly be classed feeble-minded or of questionable mentality because of deficient intellect or will (11). It is quite clear that the objective tests give a much better basis for comparison of the Bedford group with those which are to follow.

The professional prostitute confined in institutions for delinquents has been carefully studied and tested by the Massachusetts Commission for the Investigation of the White Slave Traffic, So Called (36). Three groups of 100 each were examined “without selection, except that all had a history of promiscuous sex intercourse for pecuniary gain.” One of the groups consisted of young girls under sentence in the State Industrial School for Girls, the House of Refuge and the Welcome House. A second group consisted of those just arrested and awaiting trial in the Suffolk House of Detention in Boston. The third was made up of women serving sentence in the State Reformatory for Women, the Suffolk County Jail and the Suffolk House of Correction. “These three groups represent the young girls who have just begun prostitution, the women plying their trade on the streets at the present time, and the women who are old offenders.”

The Binet tests were applied to 289 of the 300 women examined, and other psychological tests were applied in doubtful cases. The ages ranged from 12 up. Only 10 were under 15 and 32 were 36 years of age or over. The investigators classed no case as feeble-minded which did not test XI or under, but they did not class as feeble-minded 107 other cases which tested XI and under. The Commission's diagnosis is therefore conservative. It regarded 154 cases (51%) as feeble-minded, 46 in the detention house group and 54 in each of the others. If we ask how many tested below our standard we can not tell exactly, since the report does not state whether X.8 was classed as X or XI. It shows 81 tested IX or under (27%) and these were nearly all, therefore, within the limits of our group presumably deficient. Ninety-nine others tested X, a total of 60% testing below our borderline for presumable and doubtful deficients. Since only 2 cases were under 14 years of age, these figures could not be much disturbed by the younger girls. We can be reasonably sure, then, that at least 27% of these prostitutes should be placed under permanent custodial care, and probably 50% would be more nearly correct.

In a recent report of the Bureau of Analysis and Investigation of the New York State Board of Charities^[16] Dr. Jesse L. Herrick reports testing 194 inmates of the state reformatory for women known as the Western House of Refuge. The Stanford Scale was used, 25% tested IX or under with that scale and 14% tested X. In the same bulletin the report is made of Binet ages for 607 inmates of the New York Training School for Girls. Four versions of the scale were used so that the estimates are somewhat affected. Moreover, 97 girls were under 15 years of age. The table of Binet ages indicates 20% testing IX or under and 28% testing X.

Hill and Goddard (30) report examining a group of 56 girls who had been in a reformatory and were under probation with a certain officer. In this entire group they found only four who were not feeble-minded, “as we usually define feeble-mindedness.” Presumably this means three or more years retarded, including those who tested XII, so that it cannot be regarded as a conservative estimate. No further data is provided for interpreting the borderline.

Taking up the younger and milder girl delinquents, Dr. Haines reports the examination of an unselected group of 329 at the State Girls Industrial Home near Delaware, Ohio (26). They were all under 21 years of age and represent less hardened delinquents than the older groups at the reformatories for women. The Ohio group was tested with the Binet 1911 scale as well as with the Yerkes-Bridges Point Scale. Counting a result of .8 of a year as placing the case under the next mental age above, as we have in fixing the limits, we find that his results are given with such excellent detail that we may fairly compare the percentages with our standard for the Binet Scale. On this basis 70 of these delinquent girls (21%) are clearly deficient and 55 more are in the uncertain group, a total of 38%.

As a check upon results, we may compare the report of Miss Renz for 100 consecutive admissions to the same institution in 1912, tested with the Binet scale (47). She found 29 tested IX or under, 49 tested X or under, slightly more than was shown by the Haines tests. Miss Renz' report, however, does not show how many of the girls were under 14 years of age and might thus be excluded from the deficient groups.

In the California School for Girls, Grace M. Fernald^[17] examined 124 cases as they entered the school. Twenty-four tested under XI with both the Binet 1911 and Stanford revision. This is a further indication of the less frequency of feeble-mindedness in the state schools for girls than in the reformatories for women.

Dr. H. W. Crane reports the results of the Binet testing at Adrian, the Michigan Industrial School for Girls, which receives only minors and corresponds to the Ohio Industrial Home (37). The Binet 1911 scale was used, but this grouping in mental ages may mean that a few more cases are thus classed deficient than with our standardized borderlines which place the subject in the higher age group when he scores .8. It is to be remembered also that the borderlines for those whose life-ages are under 15 have not been as well standardized with the 1911 scale. The testing was done under the direction of a state commission appointed to investigate the extent of mental defectiveness (37). Dr. Crane was assisted by three other workers. The results at Adrian show, among the 386 inmates, 131 or 34% tested in our groups of presumably or uncertain intellectual deficients. Seventy-seven of these, in our uncertain group, should only class as deficient because also delinquent. The investigators give it as their opinion that 16.7% of the inmates were feeble-minded but not reached by the tests.

The entire population of the Illinois State Training School for Girls at Geneva was tested by Louise E. and George Ordahl.^[18] The Kuhlmann revision of the Binet Scale, supplemented by the Stanford Scale, for the older ages, was used. Among the 432 tested 13 per cent. tested below our borderline for the presumably deficient and 22 per cent. more in the doubtful group.

Dr. Otis, resident psychologist at the New Jersey State Home for Girls at Trenton, examined 172 girls between 10 and 20 years of age inclusive (43). Since she said it was “a preliminary testing” and “not many of the smaller girls were included,” we conclude that it was a somewhat selected group. She regarded those who stand between eleven and twelve as practically normal and those who stand below ten as without doubt defective. She then publishes three groups: “Defectives,” 45% (77 cases) high grade; “Morons,” 30% (52 cases); and “Presumably Normal,” 25% (43 cases). Since she does not give the distribution of the cases it is not possible to tell how many of her group were less than four years retarded. Her statement of the ages, however, shows that not more than 7 of the defectives could have been less than four years retarded and not more than 12 of the combined group of defectives and morons tested X or over. We may be sure, therefore, that at least 68% of these girls are of questionable intellectual ability according to the conservative standard adopted in this discussion.

Dr. Bridgman has reported the examination of 118 girls, 10 to 21 years of age, successively admitted to the State Training School for Girls at Geneva, Ill. She states that 89% (105 cases) “showed a retardation of three years or more.” The distribution of cases is not given so that it is not possible to tell how many testing X, XI, and XII were classed as feeble-minded or how many tested only three years retarded. The published estimate is undoubtedly extreme, but I have no means of making a more conservative estimate on this group. It is interesting, however, to note that only 14 of the cases were not sexually immoral. These were all cases which were either dependent or sent because uncontrollable at home and all tested as passable intellectually. She states that “according to the Binet tests, 97% of the children (5) sent to this institution because of sexual immorality are feeble-minded as well.” This percentage also would be decidedly discounted on a conservative test standard. In another place Dr. Bridgman makes the important statement that of 400 girls admitted to Geneva 60% were suffering from venereal disease (4).

Mr. Bluemel (2) found that 24 out of 50 girls sent from Judge Lindsay's Juvenile Court in Denver to the State Industrial School or the Florence Crittenden Home tested XI or under and four or more years retarded. This is less conservative than our standard, which would exclude those who tested XI as above even the uncertain group in intellect.

Dr. Pyle (46) has tested the 240 girls at the Missouri State Industrial Home for Girls with his standardized group tests. These girls are from 7 to 21 years of age and his table gives the results with each of six tests. The most significant fact for our purpose is that with the different tests from 50 to 88 per cent. fall below the averages of normal individuals who are three years younger. He says, “Our figures would indicate that about one-third of these delinquent girls are normal and about two-thirds subnormal. Most of them are probably high grade morons.” This is based apparently on 69% being the average of the results of six different tests as to the percentages three years or more retarded from their life-ages. He indicates, however, that 38%, similarly calculated, are within the average deviation of the normal groups for their life-ages. This indicates that the lowest 62% test only as low as we should expect to find the lowest 21% of random groups of corresponding ages. They should certainly not be regarded as testing feeble-minded.

(b) Women and Girl Delinquents in County And City Institutions.

When we turn to those who are cared for locally in city or county institutions, we find Sullivan (56) has examined 104 women and girls held temporarily at the Holloway jail in London, most of whom were between 16 and 25 years of age. Apparently the cases were especially selected for examination and therefore do not represent the general condition there. He was interested, however, in finding the relative amount of deficiency among different classes of these inmates and he gives the detailed results with the Binet 1908 scale on small groups of these different types which we may classify by our standard as follows:

Twenty non-criminal, either not guilty or guilty of unimportant offenses, who represent, he thinks, the ordinary conditions among the corresponding working class in this community, 3 presumably deficient, 5 uncertain; twenty criminal by reason of the occasion, 1 presumably deficient, 6 uncertain; twelve impulsive criminals, 1 presumably deficient, 2 uncertain; eight moral imbeciles, 2 presumably deficient, 2 uncertain; twenty-four recidivists, 2 presumably deficient, 8 uncertain; twenty prostitutes, 3 presumably deficient, 8 uncertain. Together these different types of women in jail form a motley group of 104 of whom 12 test presumably deficient, 31 uncertain, a total of 41%.

Ordinary prostitutes are about as frequently deficient as are those in reformatory institutions, if we may judge by an important study of women who were sex offenders but not in institutions for delinquents. The report is by Dr. Clinton P. McCord, health director of the Board of Education at Albany (35). One group consisted of fifty cases of sex offenders who were not legally delinquents at the time but were living in houses of ill-fame. Their ages ranged from 22 to 41 with an average age of 27. Nine of these (18%) tested IX or under with the Binet 1911 and 18 tested X, a total of 54% presumably and doubtfully deficient. Another 38 cases were staying at a House of Shelter where most of them had been sent by the courts. Nineteen of these tested IX or under (50%), while 13 more tested X, a total of 84%. Since their ages ranged from 12 to 40 years with an average of 18 we cannot tell how many might be above the borderline on account of an age less than 15 years, but probably very few. A third group consisted of 9 street walkers and 3 wayward girls. Among these 7 tested presumably or doubtfully deficient.

The McCord study of prostitutes not legally delinquent at the time of examination is confirmed by the Virginia State Board of Charities and Corrections in a special report to the General Assembly which gives the results of examining the prostitutes in an entire segregated district in one of the Virginia cities (58). Its table shows that, among 120 of these women, 43, or 36%, tested approximately under our borderline for the presumably deficient, while 67 cases, or 56%, tested below approximately our borderline for the presumably passable intellects.

These results are similar to Weidensall's^[19] findings among the unselected group of unmarried mothers in the Cincinnati General Hospital. While she does not give the number tested with the Yerkes-Bridges scale, she indicates that 48% tested as low-grade morons or worse, which should correspond to a test age of IX or lower. Twenty-two per cent. had intelligence coefficients of .50 or less and 32%, from .51 to .70. A Study of Fifty Feeble-Minded Prostitutes^[20] by Mary E. Paddon gives an admirable summary of the social history of prostitutes who tested deficient.

Dr. Bronner has made a careful study with Binet tests of a younger group of randomly selected girls at the Cook County Detention Home which is connected with the juvenile court at Chicago. The group included 133 girls 10-17 years of age inclusive, who were held awaiting a hearing or were temporarily cared for in the detention home. The Binet tests were given to all who did not show clearly that they were of passable mentality by completing the sixth grade or above without retardation, and passing school tests in long division and writing from dictation. A 14-year-old child “passing all the 10-year-old tests and some, but not all, of the 12-year-old tests,” was regarded as doubtful. She was not classed as feeble-minded without further testing and study. Dr. Bronner does not state her criterion for the borderline with the younger children, but we may judge that her borderline was more likely than ours to have classed a child in the presumably deficient group. Her summary shows only 15 girls “probably feeble-minded” (11.2%), and 2 others “possibly” so. From her description we may suppose that the “probable” group were comparable with our test standard of presumably deficient, plus perhaps a few conative cases.

Mention should also be made of the work of Dr. Bronner to which we referred under the earnings of the mentally retarded (6). This group of 30 randomly selected delinquent women at a local detention home in New York tested, with two or three possible exceptions, no lower than a similar group of women servants who had never been offenders. Her data do not enable us to determine how many would fall below our borderlines.

Stenquist, Thorndike, and Trabue (54) report the results with the Binet 1911 tests, under a slightly modified procedure, for 75 randomly selected dependent and 4 delinquent girls cared for by a certain county, excluding those children within the county sent to an institution for the feeble-minded. The children were from 9 to 16 years of age, with a medium age of 11 years. The line between the delinquent and dependent groups with these younger children becomes rather obscure. They state: “A child may, in the county in question, become a public charge by commitment by an officer of the poor-law on grounds of destitution, or by an officer of the courts on grounds of delinquency.... The decisive factor is often simply whether the parents are more successful in getting justices to commit their children than in getting poor-law officers to do so.” With the detailed records which they give it is possible to apply our standard even for the immature, although it is certainly less adequate for those under 15 years of age tested by the 1911 scale. I have translated their corrected Binet ages back to the original test ages, since their summary of retardation in terms of years below average ability at each age is not comparable with our borderline. Among the 79 girls who are mostly dependent, there are 5 girls, or, 6%, who fall within our presumably deficient group and 8 in the doubtful group, a total of 16%. So far as serious deficiency is concerned the situation is undoubtedly worse among delinquents than among corresponding groups of dependents. The figures of these investigators show this for their group of boys, to which we shall refer later.

Certain other groups of women and girls have been examined with the Binet or other tests, but the results are of little significance for judging the problem of deficiency objectively, since the individuals were either selected for examination because they were thought to be abnormal mentally or because there are not adequate norms for determining the borderlines with the particular tests used. At the New York State Training School for Girls in Hudson, we find that 208 selected cases who were not profiting by their training were examined with the 1911 scale. They ranged in life-age from 12 to 20. We cannot determine how many were under 14 years of age, or how much effect might have been produced by selecting dull cases; but 44 tested IX or under and 52 tested X (158). Dr. Spaulding (183) used Binet and other psychological tests on a group of 400 inmates of the Massachusetts Reformatory for Women at South Framingham; but she gives only her judgment based on the examination and history of the cases so that we have no data on this group for comparison. Her statement that 16.8% showed “marked mental defect, i. e., the moron group” and 26.8% showed “mental subnormality (slight mental defect)” is an excellent illustration of the best type of subjective judgment on consecutive cases, since she is familiar with test results. For her purpose of deciding how to care for the women it is of undoubted value, but for comparative purposes it is clear that it is impossible to tell how her subjective opinion would agree with that of an equally competent diagnostician, or what is meant by her terms “feeble-minded” or “subnormal.” For scientific purposes the Binet results for her group would be of much value, for we should like to know whether the conditions at Bedford are typical among the women's reformatories for the older offenders.

Dr. Rowland used psychological tests other than the Binet scale with a group of 35 at the Bedford Reformatory for Women, but there are no adequate norms for the comparison of her results with the general conditions (49). Baldwin (1) has shown that delinquent colored girls, 13 to 21 years of age, in the girls' division of the Pennsylvania Reformatory school at Sleighton Farm are inferior to white girls in the same institution in a learning test. As cited by Gruhle (121), Cramer (10) used an Ebbinghaus completion test, definition tests, etc. with 376 delinquent girls in Hanover, but there are no borderlines for comparison. As cited by Bronner, von Grabe gave several psychological tests to 62 prostitutes treated in the city hospital in Hamburg and compared them with a control group of 30 (6).

The most striking conclusion that comes out of the study of this evidence of frequent deficiency among delinquent girls and women is the close association between sex offenses and deficiency. One hundred and four out of 118 consecutive admissions at the Illinois training school were known to be sexually immoral. At Bedford 94 out of 100 consecutive cases had records of immorality, while three-fourths of the same group tested questionable in intellect by our standards (11). This evidence, taken with the report of the Massachusetts' Commission and the tests of sex offenders who were not at the time legally delinquents, reported by McCord, and the Virginia Commission, leaves little doubt that there is an excess of deficiency among this type of offender. Many of these deficient girls probably at first drift into the life of prostitution. They are passive rather than active agents. This distinction in the nature of the offense accounts for some of the difference between the sexes in this form of delinquency. Furthermore our public attitude in matters of social hygiene has made the isolation of the female sex more common. Part of this may be due to the greater difficulty of proof in the case of men and boys, but in part it undoubtedly means that men have not been held to as high a moral standard as women in this regard. The greater frequency of deficient sex offenders among girls, does not mean that girls are more likely than boys to be active sex offenders. They are, however, more likely to be isolated for such offenses, and also more likely to be passive offenders.

The greater amount of deficiency found among female delinquents than among corresponding groups of males is thus easily accounted for by frequent association between deficiency and sex delinquency on the part of girls and women. The combination of legal sex delinquency and deficiency is due both to a native sex difference and a difference in social attitude toward the two sexes as to this form of offense. Whichever may be the main cause of the facts found, it is clear that deficiency is, today, most serious among female offenders. It is so serious that some of our reformatories for women might even prove to be practically institutions for deficient delinquents. It is in this type of institution without doubt, that the immediate problem of the deficient delinquent is most pressing. Permanent guardianship, if not isolation, for at least a third of the inmates of an institution like Bedford which shows this amount of clear tested deficiency, under our very conservative standard, would seem to be a wise move in social hygiene. It should be undertaken at once with vigor. A more fundamental change in our social attack of this problem means state guardianship before adolescence for all girls testing presumably deficient under our standard, when their deficiency is not due to removable handicaps.

(c) Men And Boy Delinquents In State Institutions.

For the purpose of judging the importance of the question of feeble-mindedness among the most serious criminals, those committed to the state prison, we have a very important study by Rossy (48). Three hundred cases were taken at random with the exception of a few selected cases on which a report was requested. In this group, thirty prisoners could not be examined either because of language difficulties or because of their refusal to be tested. The Point Scale of Yerkes and Bridges was used and the results are presented in terms of mental ages on that scale. The examiner considered all those testing XI or under as feeble-minded and found 22% of the 300 in this class. This is less conservative than even our doubtful standard, but I estimate that 16% would fall within our doubtful and presumably deficient groups. This includes 11% who test X or under with the Point Scale plus 54% of those who tested XI. This estimate is made on the basis of the tables given by Haines (26), comparing Binet 1911 results with those of the Point Scale on the same individuals. It adds the proportion of those testing XI with Point Scale, who would test nearer X with the Binet 1911 scale.

Ordahl^[21] examined 51 convicts in the penitentiary at Joliet, Ill. They “were selected in a manner thought to secure fair representation of the prison population as a whole.” The Kuhlmann 1911 Binet scale was used and supplemented by tests for 13 to 18 years taken from the Stanford scale. It is possible that selection affected the results with this small group, since 25% showed test ages of IX or under and 36% tested X or under.

Haines tested with the Point Scale 87 consecutive admissions to the Ohio penitentiary (24). He found 18% tested below a record corresponding to X.6 on the Goddard 1911 scale, which is about the upper limit of our doubtful group.

That a smaller proportion of the state prison inmates is found intellectually deficient than is found among the inmates of the industrial schools is not surprising. This may be due to various causes. Among these may be mentioned the failure to recognize feeble-mindedness, heretofore, among the younger delinquents while the adult feeble-minded were more carefully isolated in their proper institutions. The deficient adults have also been reduced in frequency by the excessive mortality. Probably the feeble-minded are not so likely to plan or commit felony as lesser crimes and misdemeanors. Moreover the adult feeble-minded may be more stable and less inclined to delinquency than adolescents. Whatever may be the explanation, deficiency generally does not seem to be as common among the inmates of a state prison as among minor delinquents in states which are in the forefront in the care of their feeble-minded.

The state reformatories reach a class of delinquents between those of the state prisons and the state industrial schools. In Minnesota all the inmates of the reformatory except 80, who were disqualified by inability to speak English or otherwise, were tested by Dr. E. F. Green. Men are sent there only between the ages of 16 and 30, so that his table of mental and life-ages gives us the opportunity to apply our criteria accurately. Thirteen per cent. of the 370 examined tested IX or under and were presumably deficient, while 22% more were in the uncertain group testing X (22).

In a report of the Binet results with 996 inmates of the Iowa Reformatory, which Warden C. C. McClaughry kindly sent me, 200 tested IX or under and 146 tested X, a total of 35% including the doubtful group. The range of ages was from 16 to 49. The Warden notes that the tests were not made by an experienced psychologist. “In many cases it is suspected that the crafty criminal was endeavoring to lower his standing as to mentality in the hope of excusing or mitigating his crime in the eyes of the Board of Parole.” The results, however, agree well with what has been found in similar institutions.

Supt. Frank Moore of the New Jersey Reformatory at Rahway says, “Nearly every young man who has entered our institution in the last eighteen months has been tested by this system (Binet), and the results have shown that at least 46 per cent. were mentally subnormal” (38). By his discussion this seems to mean that they tested below XII which would mean that all those testing XI were less deficient than our standard for doubtful cases. These young men were from 16-25 years of age and 17.5% of them had had one year or less in school. Ten per cent. could not be examined because of unfamiliarity with English. A later report in 1912 regarding the same institution (42) says that 600 of the inmates have been examined with the Binet tests in two years, but does not state how these were selected. Of those examined we are told “48% are of the moron type of mental defectives, ranging in mentality from three to eight years, below the average normal adult.” Again, no further information is given so that it is impossible to allow for those testing X or XI or for the cases only three years retarded. Both of these estimates at the New Jersey Reformatory are excessive when judged by conservative borderlines.

Dr. Fernald has applied 11 objective tests to a representative group of 100 inmates at the Massachusetts Reformatory (15) but the norms for the tests which he used were obtained, for the most part, by testing a dozen boys so that the line which he draws for the limit of the defectives is largely a matter of his expert opinion and the estimation loses objective character. He estimates that 26% of his group whose ages run from 15 to 35 inclusive were defective. Beanblossom^[22] has published an account of tests on 2000 inmates of the Indiana Reformatory. Some of the Binet tests as well as other tests were used but the published results do not admit of reinterpretation.

Comparing the reports from the Minnesota, Iowa, and New Jersey reformatories with the tested deficiency found in institutions for women delinquents on the basis of the same borderline with the scale, the records indicate clearly that the percentage of feeble-mindedness is greater in the reformatories for women. At the Bedford Reformatory for women, for example, Dr. Weidensall's results show that the corresponding borderline to that used in the New Jersey men's reformatory which reported 46% deficient, would class 100% at Bedford as feeble-minded, where only one case in 200 tested as high as XII. A conservative estimate of tested deficiency in men's reformatories from the above data would be from 15 to 20%.

In the state institutions for minor delinquents, usually called industrial schools, we have several studies of representative groups with sufficient data to make objective interpretations comparable with our standard. In Ohio, Dr. Haines (26) reports on the examination of 671 delinquent boys 10 to 19 years of age at the Boys' Industrial School near Lancaster. Interpreted as we have indicated for the Ohio Institution for girls, we find 100, or 15%, in the group testing presumably deficient and 179 in the doubtful group, a total of 42% clear and questionable.

In the corresponding Michigan Industrial School at Lansing, Dr. Crane (37) shows by his table of mental and life-ages that 52 out of the 801 unselected inmates, or 6% are presumably deficient and 171 below the presumably passable, or 21%. This is only a slightly greater number than our criterion would provide, if .8 of a year were not classed in the next higher mental age by these examiners. The age of those examined ran from 10 to 17.

T. L. Kelley in his “Mental Aspects of Delinquency”^[23] gives the results for an extensive series of measurements and tests on about three hundred boys in the Texas State Juvenile Training School. On the basis of an analysis of his tests he estimates that 20% of the boys there should be in a school for the feeble-minded. Interpreting his original data for the 1911 Binet tests on the same basis as our own, 8% fall within the clearly deficient group and 9% in the doubtful. The latter on account of their delinquencies might also be included as feeble-minded.

The 215 inmates of the Whittier State School in California were examined by J. Harold Williams with the Stanford revision of the Binet scale (61). The boys were 10 to 22 years of age, median 16 years. He states that 32% were feeble-minded in the sense of having Intelligence Quotients less than .75. This is a standard which would include about 2% of those tested with the scale, so that we may consider the bulk of them as within our presumably deficient and uncertain groups combined. He also states that approximately 14% tested below X with the Stanford Revised Scale. In another paper he shows that the amount of feeble-mindedness was much different among the different races represented in the institution. With 150 cases according to his standard there were 6% feeble-minded among the whites, 48% among the colored, and 60% among the Mexican and Indian races. In this group 64% were native whites, 21% of Indian or Mexican descent and 15% colored. “While the negro population of California constitute but 0.9% of the total, yet the results of this study indicate that more than 15% of the juvenile delinquents committed to the state institution are of that race.” It is, of course, of fundamental importance in regard to all estimates of feeble-mindedness among delinquents to consider the racial conditions at the particular institution.

A New Hampshire Commission tested the children in its State Industrial School. Its table shows that among the 113 boys tested at least 37% were presumably or doubtfully deficient. To these should be added some 14 years of age and over who tested X, in order to have the total number below our borderline for the presumably passable cases. The published table does not separate these from the 13-year-olds (40). Hauck and Sisson report in School and Society for September, 1911, tests made at the Idaho Industrial School, which receives both boys and girls from 9 to 21 years of age, including some children who would be classed as dependents but can not be cared for elsewhere in the state. Supposing that our standard applied to the 1911 scale which was used, among 201 tested there were 5 presumably deficient and 13 doubtful.

A partially selected group of 341 inmates at the St. Charles, Ill., State School for Boys chosen in such a way that it naturally would somewhat increase the frequency of deficiency, was tested by Dr. Ordahl with Kuhlman's form of the 1911 scale supplemented by the Stanford Scale above XII. The results showed 11% in the presumably deficient group and 20% in the doubtful group (41).

One of the main uses of the objective scale is to demonstrate that the same conditions do not prevail in various institutions which, except for this objective evidence, might be expected to care for the same type of inmates. This is illustrated by the comparison of the above studies in Ohio and Michigan with that made at a similar state school for delinquent boys in Indiana reported by Hickman (12, 28). The Binet 1911 tests, Goddard's adaptation, were applied to 229 new boys 8 to 17 years of age inclusive, admitted to the Indiana Boys School at Plainfield. Among these, 68 boys (30%) tested below our borderline for the clearly deficient and 53 more within the doubtful region, a total of 48%. There seems little doubt that this represents a significant difference from the condition at the corresponding Ohio and Michigan schools where only 15% and 6% respectively tested clearly deficient on a corresponding standard. An interesting commentary on the necessity of reinterpreting the borderline for feeble-mindedness on the scale arises when we note that Hickman says: “One hundred and sixty-six, or about 75% of the whole number tested, tested as much as three years or more below normal, and therefore would be classed as feeble-minded to a greater or less degree.”

(d) Men And Boy Delinquents In County And City Institutions.

It seems likely that in city and county institutions deficiency is most common among repeaters in the jails or workhouses. One study has been made of a randomly selected group of repeaters who were in the jail of a Virginia city for fixed sentences of not more than a year. The examinations are summarized in the Special Report of the Virginia State Board of Charities and Corrections (58). In this Virginia city 50 whites of both sexes and 50 negroes of both sexes were examined. Among the whites, 18 tested IX or under and 5 more tested X. Among the negroes, 24 tested IX or under and 10 tested X. The percentages would be just twice these numbers, a total of 61% below passable capacity in this group of 100. If such is the condition in other jails in other parts of the country, it indicates one of the most serious hot beds of deficiency among delinquents. The repeaters in this city jail during three years were responsible for 60% of the commitments to jail, although only about one-fourth of the 33,306 arrests in this city during the three years resulted in commitment to jail. The feeble-mindedness among the repeaters, therefore, may be little indication of the frequency of deficiency among those arrested in the city. The repeaters represented only a third of those committed to jail during this period and this third was probably the most deficient among those committed, since recidivism goes with deficiency. Moreover, those committed to jail are probably more likely to be deficient than those who escape jail sentences. To assume, therefore, that 61% of this city's delinquents were of doubtful ability would be clearly unjustified, and yet this sort of reasoning about the frequency of deficient delinquents has been all too common.

Gilliland^[24] tested one hundred male inmates of the Columbus, Ohio, Workhouse (28 negroes) selected so as to attempt to represent the different offenses about in their proportions. He gives the results in point scores with the Yerkes-Bridges scale, which may be translated only roughly into Binet 1911 ages by Haines' data, as I have indicated for the study by Rossy. All were 18 years of age or over, so that I estimate 14% would fall into our presumably deficient group including only the proportion of those under 64 points who would test as Binet IX or less. The doubtful group would include 17% more, including the proportion under 66 points who would test X or under.

Among the local institutions supported by the county or city, the most serious delinquency is probably found in the group reported by Kohs at the Chicago House of Correction (33). He tested with the 1911 Binet scale 335 consecutive cases between 17 and 21 years of age. Among these were 72 cases (21%) who tested clearly deficient according to our standard, and 95 cases doubtful, a total of 50% at least uncertain in intellectual ability.

Through the courtesy of Catherine Mathews, who made the examinations for the psychological clinic of the University of Pittsburgh, which is under the direction of Dr. G. C. Bassett, I am able to give the records of 125 consecutive admissions to the Allegheny County Detention Home. The institution is known as the Thorn Hill School. It is situated some miles outside of Pittsburgh and provides on the cottage plan for about 300 boys. The boys are sent from the Juvenile Court for milder training than that at the state school. The school has also been found to furnish a necessary place to care for cases of feeble-minded delinquent boys who cannot be immediately admitted to the state institution on account of its crowded condition. A detention home is also provided in the city for juvenile court children awaiting trial or the disposition of their cases. These are not included in the Thorn Hill group.

Among the 125 consecutive cases at Thorn Hill, omitting two cases which are probably dementia praecox, there were 37, or 29%, who tested presumably deficient according to our standard, and a total of 68 cases, or 55%, presumably and doubtfully deficient. It is to be remembered that our standard for the immature was arranged for the 1908 scale and not the 1911 scale which was used here, although the difference would be slight.