The Intelligence of Policemen

Louis L. Thurstone

Carnegie Institute of Technology, Pittsburgh; Pa.

The present paper is the result of a study of the Detroit police force undertaken by Dr. L. L. Thurstone of Carnegie Institute at the invitation of the Detroit Bureau of Governmental Research. The data for Detroit are compared with those obtained from a similar study in Cleveland conducted by the Cleveland Foundation.

Summary.--Police lieutenants in two cities are found to have on the average less intelligence and less schooling than the patrolmen whom they command. Length of service correlates negatively with intelligence as measured by the Army Alpha examination. The most probable explanation of these disconcerting facts is found in the relatively greater attractiveness of other occupations, into which the brighter and better educated patrolmen transfer, rather than wait for promotion. Readers of this report will recognize the bearing of these facts on problems not only of the civil service, but also of all departments of industrial concerns in which promotion is delayed and recognition of special merit is slow. If the ablest employees are to be retained, advancement must be sure and fairly rapid.

LAST summer I had the opportunity to make an investigation of the intelligence of policemen in the city of Detroit and I shall report in this article some of my findings not only with reference to the intelligence tests but also with reference to the interpretation which these records give on some personnel problems of the police service as an occupation.

My relations with the police department of Detroit were through the Detroit Bureau of Governmental Research. I am indebted to Dr. Lent D. Upson, director of that bureau, and to Mr. Arch Mandel of its staff for the opportunity to conduct this study.

Previous to this study the appointments to the police department had been made by civil service examinations prepared and conducted by the civil service commission. Owing to some disagreements regarding the effectiveness of these examinations the police department expressed a preference to give their own examinations for appointment to the police service. At this time I was consulted on the question as to whether any of the psychological procedures with which I have been working could be applied in the police service. I could not say without a try-out whether psychological tests could be applied in this new field. It was therefore decided that I

( 65) should prepare a series of tests and that these tests should be given to a large number of policemen of varying ranks in the Detroit Police Department. The test records were to be compared with the service records-of the men who took the tests in order to determine whether the tests have diagnostic value. If the tests should prove useful in such an investigation they could be used in the future for purposes of selection and appointment to the police department. To this program I agreed and I prepared several examinations for the purpose.

As a test of intelligence I selected as most suitable the Army Alpha Intelligence Examination. It has been extensively used already for men whose intelligence is more or less comparable with that of policemen and it has also been used recently in Cleveland to measure the intelligence of the police-men in that city. To use the same examination makes it possible to make comparisons which I shall report here. In addition to this examination I gave a test of immediate memory consisting of the picture descriptive of a street car accident which is included as part of the tests of journalistic ability in Mr. Freyd's series. This immediate memory test was given with a separate time limit and a separate score. I also used a personal history record and a list of questions bearing on social and emotional traits which I called a self analysis.

At the present writing I have analysed the raw scores sufficiently to feel fairly sure that I shall not be able to find anything strikingly diagnostic in any of the tests. None of the tests that I used and none of the personal history items and self analysis questions seemed to differentiate the men -who have low service records from those who have high service records. I shall report briefly in a separate paper my findings in all of the tests with special reference for diagnostic value for the-"police service with service records as a criterion. My present purpose is to analyse the Army Intelligence Test records in the Detroit and Cleveland police departments and to interpret certain discrepancies that are conspicuous in both cases.

When the papers had all been marked and tabulated my first curiosity was to compare the lieutenants, sergeants, and patrolmen in Detroit on the Army Intelligence Test. I expected the officers to score higher than the patrol-men but the following is what I found.

Detroit Police Department   Number of Men   Average Alpha
Intelligence Scores
Lieutenants   17   57.80
Sergeants   34   54.71
Patrolmen   307   71.44

Inspection of these averages reveals immediately that the patrolmen score on the average noticeably higher than the officers in the Detroit department. One would expect the officers to score higher than the patrolmen because the officers are selected supposedly from among the brightest and most capable patrolmen. This we do not find in the above table nor in the tabulations that follow in this report.

We do notice however from the above table that the lieutenants score higher than the sergeants. This is what one would expect since the lieu-tenants are, supposedly, on the

( 66) average a more select group than the sergeants.

With this finding one is challenged to discover a reasonable cause. Most of the patrolmen who were examined in this investigation were appointed during the last year or two. I have listed` four possible reasons for this peculiar discrepancy and I shall recommend one, of these explanations as the most prominent cause.

1. It may be that the Intelligence Test does not measure the type of ability which marks a man for pro-motion in the Detroit Police Department. This could conceivably be the case if personality and character traits, volitional traits, and mental stability were more important for success in the police service than general intelligence. Another logical interpretation would be that the test does not measure intelligence. This possibility is excluded with reference to the Army Alpha Intelligence Test because it has in so many other occupations demonstrated differentiating value with regard to success and failure with those occupations and estimates of general intelligence,

II. It may be possible that efforts have been made during the last year or two to recruit men of higher mentality for the police service than formerly. If that should be the case, then it would be only natural to expect as a consequence that the men who have recently joined the service and who have not yet been promoted beyond the rank of patrolman should score higher than those who have remained for many years in the service and who have in consequence attained the officers' ranks,- If this explanation is adopted the finding in the above table could be interpreted as a compliment to the current administration for its efforts to get men for the police service of higher mentality than formerly.

III. Another possible explanation of the higher intelligence scores of the patrolmen would be in connection with the recent industrial depression which has thrown out of employment thousands of high grade mechanics. It may possibly be that a certain percentage of high grade men who are are temporarily out of employment may have sought entrance to the police department on account of the civil service stability of that service as an occupation. If this explanation is accepted the higher rank of the patrolmen who have entered the service within the last year would be a temporary effect which could not be relied upon for the future personnel of the department. It may be possible to check this explanation by analyzing the vocational experience of the men who have entered the service during the last year and of those who entered some eight or ten years ago.

IV. Still another explanation of the high score of the patrolmen compared with the intelligence scores of the officers might be found in the nature of the police service as an occupation. I have reference to the possibility that the police service is such an occupation that men of superior mentality do not find it congenial-and profitable to remain in it. If we assume that a certain sampling of the population enters the police service with an average intelligence score of 71, and if we assume that the brightest men leave the

( 67) service sooner or later in pursuit of other occupations, then it would be only natural to expect that those who remain in the service to be promoted would average lower on the Alpha Intelligence Test.


I consulted the Memoirs of the National Academy of Sciences; Volume XV, for intelligence test records of policemen from different parts of the country. These records were obtained by classifying the intelligence test scores in the draft army by occupations. The records in the following table are obtained from Chapter 15, "Intelligence Ratings of Occupational Groups."

Policemen's scores in Army Alpha Intelligence Test
  Number of Cases   Alpha Average
Camp Wadsworth 24   119.0
Camp Devens 26   139.0
Records in Surgeon General's
Office in Washington
107   69.3
Detroit Lieutenants 15   57.8
Detroit Sergeants 34   54.7
Detroit patrolmen 307   71.4
Detroit, three ranks combined 358   69.0

The records from Camp Wadsworth and Camp Devens give the intelligence test scores for small groups of men drawn from the police service of various cities. The Alpha average of the twenty-four policemen at Camp Wadsworth was 119. The same average for the twenty-six policemen represented in the records from Camp Devens was 139. These records are unusually high and indicate that the men at these two camps who gave the police. service as their previous occupation scored unusually high in the intelligence tests. The samplings are, however, small and the averages are therefore not especially reliable.

Another more reliable source is the Army Alpha average for the 107 police-men found in the occupational classification of certain large groups of the draft army whose records are on file in the Surgeon General's office in Washington. Their Alpha average is 69.3. These records show that the Detroit lieutenants and sergeants score unusually low in intelligence rating when compared with other available intelligence ratings of policemen. It is interesting to note that the average intelligence rating of Detroit patrol-men (71.4) is equal to or but slightly higher than the general average intelligence rating for policemen in the Surgeon General's records. This would indicate that the present patrolmen in Detroit are more or less typical of their occupation as far as intelligence rating is concerned, and that the Detroit police officers are noticeably inferior to the general run of these records.


Since complete records are available on the intelligence tests for the Cleveland Police Department, it is interesting to take the opportunity to compare the intelligence averages in these two cities. I have drawn these Cleveland data from Parts III and V of the reports of the Cleveland Foundation. The reports are entitled

( 68) "Criminal Justice in Cleveland". In the following table, I have reproduced the median scores of the Cleveland police. I have also indicated the median scores of the Detroit police of corresponding rank so that direct comparison may be made between the intelligence medians of the two cities for equal ranks.

Rank or Division Cleveland
Captains 98 C plus  
Lieutenants 95 C plus 61 C
Sergeants 99 C plus 58 C
Vice Squad 75 C  
Detectives 59 C  
Training School 63 C  
Traffic 61 C  
Mounted 78 C plus  
Emergency 67 C  
Patrolmen 67 C 71.4 C

In the above table we note that the officers in the Cleveland police force score much higher on the Army Alpha than the officers in Detroit. Compare for example the Detroit average of 61 for lieutenants with the corresponding average of 95 for Cleveland officers of the same rank. Compare likewise the Detroit average of 58 for sergeants with the Cleveland average of 99 for officers of the same rank. We note, however, that the Detroit average for patrolmen is slightly higher than the Cleveland average. Compared by medians, these are 71.4 for Detroit and 67 for Cleveland.


The discovery of the low intelligence scores of the Detroit officers compared with the intelligence scores of the patrolmen in the same department led to further inquiry about the mentality of the men who are promoted.: We note, then, that in Cleveland the officers score very much higher in the Army Intelligence Test than the patrolmen. This is what one should

chart 1

expect because the officers are, supposedly, selected for superior ability and intelligence. I have represented in Chart I, the average intelligence test scores of the three ranks,

( 69) lieutenant, sergeant, and patrolman for the Detroit department. On the same chart, I have also represented the average of the last grade of school attended. This shows that the Detroit patrolmen had an education indicated by the average of 7.56 grades of grammar school which is higher than the average schooling of the officers. The chart also shows that the Detroit sergeants had less schooling on the average than the Detroit lieutenants. This is what one should expect in the tabulation of average education. It is interesting to note that the order of the three ranks—patrolman, lieutenant, sergeant is the same when we compare them on the basis of general schooling as when we compare them by the Army Alpha Intelligence Test. This similarity is quite conspicuous in Chart I.

This analysis of average schooling of the three ranks verifies the inference that one would draw from the intelligence test alone; namely, that the Detroit patrolmen are as a group superior in ability and intelligence to the Detroit lieutenants and that the lieutenants are superior in the same respects to the Detroit sergeants. Both the intelligence test and the average schooling show the same fact.

The data in the National Academy Memoirs to which I referred also contain information regarding the average Alpha score and degree of advancement in each occupation. It is interesting to note that in several occupations the intelligence test averages decrease with advancement and experience in the occupation. In the majority of the occupations the intelligence test scores increase with recognized advancement. This is what one should expect because other things being equal, the man of superior general ability and intelligence is the one who rises to the ranks of journeyman and of expert in his occupation. When we find an occupation in which the higher ranks show lower intelligence test scores than the apprentices in that occupation, we are justified in assuming that the occupation does not appeal to the brightest apprentices. We are forced to assume in most of these cases that the brightest apprentices leave the occupation for some other line of work in which their superior ability can be more profitably used. This is what I suspect to be the case with the Detroit Police Department on account of the conspicuously low intelligence test scores of its officers.

The records of the Surgeon General's office are kept in uniform terminology with reference to the several degrees of advancement in each of the many occupations that are listed. Their uniform nomenclature for occupational advancement is indicated by the terms: novice (n), apprentice (a), journeyman (j), and ex-

pert (e). This terminology is of course common in many of the trades but it is not commonly used in some other occupations. It does, however, serve to indicate degrees of advancement in the occupation even though the terminology be different as it is in the case of the police service as an occupation.

( 70)

Consider now the following table:

Army Records Alpha Average Alpha Average Detroit Records
Apprentice Rating 77.3 71.4 Detroit Patrolmen
Journeyman Rating 69.0 54.7 Detroit Sergeants
Expert Rating 64.0 57.8 Detroit Lieutenants

In this table I have condensed the averages for the three successive ranks of occupational promotion in the police service. These records show that for the available data the apprentices' average rating in the Army Alpha Intelligence Test is 77.3. This is compared with the intelligence average of the Detroit policemen. The correspondence may not be absolutely perfect but the table does show that as we increase in rank in the. police service, the intelligence test averages de-crease. Those men in the draft army who were by their experience and previous positions rated as experts in the police- service had obtained an average intelligence test score of only 64.

The point that I want to make clear is that the lower intelligence test scores of officers in the police service is not individual for the Detroit department but that it is indicated also in other available records about this occupation. These data at least point to the possible conclusion that the brightest men who enter the police service dropout from that occupation in favor of other occupations where their ability is better recognized.

Consider now, Chart II. In this chart I have tabulated the average intelligence test score against the number of months in the service. The conspicuous fact is that those men who have been in the service for a number of years are on the average less intelligent than those who have recently entered the service. The last three columns in Chart II show also very clearly that among those who have been in the service twenty or more months, the less intelligent have apparently been promoted to officer's rank.

I was curious to discover whether such a relation existed also in the records of the Cleveland Police Department. I have made a similar tabulation in Chart III. The coordinates differ slightly on account of the manner in, which the data were given for Cleveland. I have tabulated graphically the per cent with superior intelligence and the per cent with inferior intelligence against number of years Ca the service. This chart shows in a striking manner that the bright men leave the service. In the Cleveland department 27 per cent are of superior intelligence among those who have entered during the past year. Only 5 per cent of those who have remained seventeen or more years are rated as of superior intelligence. When the brighter men leave the service sooner or later, it is only natural to expect that the percentage of inferior intelligence should increase with years of service. This is what actually hap-pens as seen in Chart III.

( 71)

Chart 2

Chart 3

( 72)

There are more than five times as many men with superior intelligence in the newcomers of the Cleveland department as among -those who have been in the service seventeen years or more.

chart 4

In Chart IV, I have made a similar analysis of the percentages of superior and inferior intelligence with reference to length of service in the Detroit Police Department. This chart brings out conspicuously the same fact as was found for the Cleveland department, namely, that the percentage of men with superior intelligence decreases with years of service. Corresponding to this loss of the brighter men, we have an increase in the percentage of men of inferior intelligence with years of service.

The objection might possibly be raised to my interpretation of these charts in that the older men may score lower on the intelligence test on account of their advanced age. Some effort might then be made to show that the decrease in the intelligence test scores with years of service would be caused by. advanced age. In Part V' of the reports of the Cleveland Foundation Survey, previously mentioned, I find the following statement: "Table 4 shows a distribution of intelligence ratings of patrolmen according to the date of entry into the police department. There is very little difference in the ratings of the first year groups, who entered between 1895-1919. There is a slight decrease with length of service, much of which may be attributed to the deterioration of increasing age." This conclusion is not sound. Look at Charts III and IV. You will note the striking differences in the percentage of police-men with high intelligence for the groups that have been in the service less than one year, two years, and three years. It would be absurd to claim that the Alpha intelligence score is reduced by the factor of senility with-in two years of appointment to the service! Hence I interpret these data as indicative of the fact that bright and capable men do not remain in the police service as a career.

( 73)

Another possible explanation of the high percentage of superior men among those who have recently been appointed could be offered with reference to the recent industrial depression. It might be argued as a logical possibility that unemployment during the last year is responsible for a sudden influx of men with superior mentality into the occupation of the police service. This conclusion would be justified if the intelligence test scores were found to drop suddenly for one year of service and if these Alpha averages were found to remain fairly `constant from then on. The effect to which I am here calling attention is, however, too uniform in Cleveland and Detroit as well as in the corresponding tabulations made for police service advancement which was attained before the war. The recent considerations of unemployment could not be offered in explanation of the similar findings in the Surgeon General's records which represent trade advancement attained before entry into the draft army.

I believe, therefore, that I am justified in suggesting that the police service as an occupation be analyzed with special reference to the opportunities that it offers for bright and capable men. Can something be done to in-duce men of superior mentality to seek the police service as a profession and a career and to remain in it? If the social functions served by the police department are dignified as worthy of the status of a profession, it should be possible to retain in that service the brighter men who are now found to leave it.

It seems to me that the police service as an occupation should be divided more or less on the plan of the army with reference to officers, non-commissioned men, and privates. One cannot expect the educated and professionally trained men to spend five to ten years in uncertain apprenticeship, walking a beat or guiding traffic, before reaching the more distinctly professional work of the police service. On the other hand, one. can-not expect that those whose education and abilities find occupational satisfaction in the work of a patrolman should be competent to rise by promotion into the more responsible positions in the service. Would it not be better to divide the service into two classes; the work of the patrolman, and the work of the officer, and to in-duce educated young men to enter the police service to become officers after a short period of apprenticeship and a relatively short period of specialized instruction? Obviously such a system would not satisfy democratic ideals unless there was a provision made whereby a bright and promising patrolman might become an officer by acquiring the necessary education.

The minimum education required of police and detective officers should be a high school education and one or two years of specialized training in the professional methods of police and detective work. It might even be possible to arrange for instruction in some of the social sciences related to the work of a police officer such as, sociology, municipal government, social hygiene, psychology, statistics. It would probably be impossible to recruit officers competent to work out their problems in an intelligent and a professional way if we limit the selec-

( 74) -tion of such officers from among those men who have been willing to spend four or five years of their lives as patrolmen. The man who has the limitations in mentality and education requisite for contentment in the work of a patrolman cannot be expected to see problems of crime in the large. He cannot be expected to develop professional competence in the intricacies and ramifications of dealing with crime. Effective leadership in this line requires first of all a good general education and considerable professional training in methods.

Even if the police officers are recruited from high school graduates or .even from among college graduates, there should be nothing to prevent an unusually capable patrolman from attaining the rank of officer if he can ac-quire. the necessary education.

To install a plan along this general line would naturally meet with immediate opposition for political reasons, but that should not prevent us from taking steps that lead in this direction even though it would take ten years or more to put this plan completely in operation.


No notes

Valid HTML 4.01 Strict Valid CSS2