Thirty-five years ago, when Sheldon Gluecks suggested to his professor at the GSAS that an evaluation be made of the peno-corrective system used in this country, he had no idea that he was opening the door on an area of research that would become a lifetime career for him and his wife, Eleanor. His instructor seized upon the suggestion, secured $3,500 for research, and in 1925, the Gluecks began an unprecedented search into the facts and fallacies of the peno-corrective system of the United States.

This week they published their 12th book, Predicting Delinquency and Crime, a volume containing over 50, unique prediction tables to help judges, prison officials, social workers and clinicians. These tables are based on a series of statistical surveys, begun in the twenties and presented in the couple's first three books--500 Criminal careers; 500 Delinquent Women; and 1,000 Juvenile Delinquents.

"Each of these original studies uncovered facts that were more and more disturbing to more and more people," Mrs. Glueck, now a research associate at the Law School, said recently. For, instead of finding that the penal system was turning out permanently reformed criminals as was expected, the Gluecks discovered that about 80 per cent of the adult male offenders released, continued to disobey the law and about 88 per cent of the juvenile delinquents recidivated within five years after their dismissal.

Society Neglects Early Dangers

Perhaps the most important fact reaped from the Glueck's study of adult offenders was that 75 per cent of them had also been juvenile delinquents. This information inspired the couple to begin their famous, very extensive study into the causes and prevention of juvenile crime. Their work has been sponsored as a permanent project by the Law School, where Glueck is Roscoe Pound Professor of Law.

The Gluecks have gradually become convinced that society neglects the early danger signals of personality maladjustment and of criminality by not providing adequate means of screening-out children before they become overt delinquents and by not providing satisfactory school clinic facilities for study and treatment of problem children.

The couple's most famous study, published in Unravelling Juvenile Delinquency, (1950) was their first project done with a control group of non-offenders. Five hundred delinquents and 500 non-delinquents ranging in age from 11 to 17, were matched case for case by age, residence in underprivileged areas, ethnic origin, and intelligence level. The Gluecks systematically compared 402 factors in the youths' family and home backgrounds, school history, leisure-time and interests. Their development health history, physical condition and body structure, underlying personality, temperamental traits, and the quality of their intelligence were also recorded.

The delinquents as a group were found to differ markedly from the non-delinquents in five major ways: socioculturally, temperamentally, in attitude, psychologically, and physically. Socioculturally, the offenders had been reared in homes of little understanding, affection, stability, or moral fiber by parents usually unfit to be effective guides or protectors. Temperamentally, the delinquents were more "restlessly energetic, impulsive, extroverted, aggressive, destructive, and often sadistic." In attitude, they were far more hostile than the non-offenders, far more "definant, resentful, suspicious, stubborn, socially assertive, adventurous, unconventional, and non-submissive to authority."

Psychologically, the delinquents tended more to direct and concrete, rather than symbolic, intellectual expression and were less methodical than the non-delinquents in their approach to problems. In physical make-up, the offenders were essentially mesomorphic (solid, closely knit, muscular).

These discoveries disproved several popular fallacies about the causes of juvenile delinquency; for example, the idea that delinquents are physically unhealthy children. The Glueck's findings show that, if anything, the delinquents were in better health than the non-delinquents: 91 per cent were rated in good health as compared with 88 per cent of the non-delinquents, according to a standard medical examination.

The Gluecks also disproved the idea that glandular disturbances cause delinquency, since their staff physician found such disturbances among 32.9 per cent of the offenders and 34.3 per cent of the non-offenders. They also showed that delinquents are not more emotionally unstable than normal children. In fact, their study revealed that there was emotional weakness among 49 per cent of the delinquents as contrasted with 56 per cent of the non-delinquents.

Another fallacy that the couple disproved by this extensive project, is that large families are the cause of delinquency. As it happened, both groups of families were equally large, averaging six to seven children.

The most important fact to come out of the study for Unraveling Juvenile Delinquency, was that half of the delinquents had been less than eight years old when marked signs of their antisocial tendencies first became evident, and nine-tenths of the 500 had manifested these tendencies before reaching 11 years of age. This new knowledge, coupled with the previous discovery that about 75 per cent of adult offenders had been juvenile delinquents, made it clear that community effort should be directed toward very young children, those less than seven years old, if juvenile delinquency is to be reduced significantly.

Three predictive tables were formulated from the findings in Unraveling Juvenile Delinquency. These tables were designed to make it possible to spot possible delinquents at a stage when they can be steered away from a life of crime. Of the three devices, only one, the Social Prediction Table, has been validated since its inception in 1950. The other two, which are constructed from Rorschach and psychiatric data, are not easily applied, and together do not do a better job of selecting potential delinquents than the SPT does alone.

The SPT is composed of five factors with subcategories (see box). The scores for each subcategory represent the percentage of delinquents among the 1,000 cases of Unraveling Juvenile Delinquency who fell into the particular subcategory. For example, if a particular child has been harshly disciplined by his father, he is scored 72.5, because 72.5 per cent of the boys in the study whose fathers were always or sometimes over-strict in their discipline, were found to be delinquent. If a child scores 300 or more in the Social Prediction Table, his chances of becoming a delinquent are high.

The first problem faced by the Gluecks after setting up this predictive table was to encourage its application to juvenile delinquents with backgrounds differing from those who made up the original group. If offenders had the same "cluster of factors" as were found among the SPT delinquents, the test would be validated. The process was then to be reversed--the test applied to six year olds to see if predictions of delinquency actually came true.

Several retrospective validations of the Social Prediction Table have been made since its publication in 1950. The first appeared in the spring of 1952--a study made by Bertram J. Black and Selma J. Glick of the Jewish Board of Guardians in New York City.

The table was applied to a group of 100 Jewish boys confined in the Hawthorne-Cedar Knolls School in New York State, with the purpose of determining the extent to which it would have been possible, years earlier, to have identified them accurately as potentially serious delinquents. Black and Mis Glick ascertained that 91 per cent of the group would have been correctly spotted by the test.

It was encouraging to the Gluecks to find that although the SPT was compiled on the basis of underprivileged Boston boys largely of English, Italian, and Irish descent and of Protestant and Catholic religions, it operated so well on a sample of New York Jewish boys.

Another study was made by Richard E. Thompson '52, which he submitted as his Senior Honors thesis in the Department of Social Relations. Thompson established the SPT as a valid instrument for distinguishing among children already showing behavioral difficulties, those who are true delinquents from those whose maladapted behavior is temporary. The SPT showed that among a representative group of 100 boys, included originally in a research project called the Cambridge-Somerville Youth Study, it would again have been possible to identify accurately 91 per cent of all the boys as either potential delinquents or as non-delinquents.

The discriminative potential of the SPT was found to be considerably greater than that of three clinicians (a psychiatrist, psychologist, and criminologist who had been initially charged with selecting the boys for the Cambridge-Somerville Youth Study. Thompson reported that the clinicians had correctly identified only 65 per cent in comparison with 91 per cent correctly identified by the SPT.

In this inquiry, again the table revealed its value for boys of status and background different from that of the original test group. The Cambridge-Somerville Study included boys of higher intelligence, of better economic status, and of better home neighborhoods than those in the Unraveling Juvenile Delinquency study.

Still another check on the SPT was published in April, 1955, by the Jersey Department of Institutions and Agencies in which the table was applied to 51 delinquent boys who were on parole. The group reports that "the closeness of our findings with the original findings in the study of Unraveling Juvenile Delinquency is rather noteworthy, since the New Jersey boys were selected at random, and no attempt was made to match the individual characteristic of these delinquent boys with those included in the Harvard Law School study."

In addition to these published projects several other checks have been made of the table. Thompson, in a second test, applied the SPT to 50 boys who had appeared before the Boston Juvenile Court in 1950.

Working retrospectively, he found that if the table had been applied when these boys were six years old it would have predicted that 92 per cent of them were destined to become delinquents.

Another test of the SPT's validity was made in 1954, when the Douglas A. Thom Clinic for Children in Boston, applied it to 57 boys ranging in age from 6 to 12 years, who had been treated for aggressive, destructive, anti-social behavior. The scorings made by the clinic psychologist indicated that 823 per cent of these boys, at the age of six, would have been clearly identified by the test as potential offenders.

Thompson, in a third experiment, applied the table to 50 girls, committed to the Division of Youth Service of the Massachusetts Department of Education between November, 1954, and May, 1955. This was only the second application of the test to girl delinquents. The first had been done by the Jewish Board of Guardians, in 1953, on 150 unmarried mothers, whereby it was found that 81 per cent of the women would have been correctly identified at the age of six.

The result of the application by Thompson to girl delinquents is unusual since all of the 50 girls would have been identified as potential offenders. Mrs. Glueck offers the explanation that the social pathology in the background of delinquent girls is far worse than in the background of boys. "Girl offenders are not brought to court unless they are really, seriously delinquent," she said.

This was clearly reflected in the categorization of this group of girl offenders on the five social factors that comprise the prediction table. Laxity of discipline by father, for example, was found in 75.6 per cent of the girls as contrasted with only 26.6 per cent of all the boys in Unraveling Juvenile Delinquency. The fathers of the girls were indifferent or hostile to them in 87.5 per cent of the cases, as compared with only 59.8 per cent of the boys.

An application of the SPT to 150 delinquent boys in upper income groups ($7,500 and thereabouts) by the Jewish Board of Guardians is still in process. Thus far, the table has been applied to 81 boys and reports state that it would accurately have identified 89 per cent of these.

Some critics claim that the groups tested with the SPT have been too small to be definitive. The Glueck's, however, feel that since all the experiments have pointed in the same direction, it must be concluded that "we are on the track of a useful device."

It seems a reasonable assumption that the appearance of the SPT factors in the background of children not yet delinquent is a danger signal of budding criminal careers. This is presently being rigorously tested.

Two studies are in progress, one in New York City and one in Washington, D.C. The first is being carried out by the New York City Youth Board, applying the table to about 250 boys in the first grade of two public schools in high delinquency areas in the Bronx. This experiment has been in process since 1953, and will be continued until the youths are about 17 years old, in 1964.

In an interim report published in 1957, the Youth Board said that the