Some Fukther Considerations Upon the Re-Tardation of the Pupils of Five City School Systems
The Psychological Clinic Vol. II. No. 3. May 15, 1908. :Author: Roland P. Falkner, Ph.D. Commissioner of Education for Porto Rico, 1904-07.
In the February issue of The Psychological Clinic, Dr. Oliver P. Cornman presented a valuable discussion of the Retardation of Pupils of Five City School Systems.* His study is based upon tables showing the distribution by ages of the pupils in each grade of the elementary schools of the cities of Camden, Kansas City, Boston, Philadelphia, and New York. The text and tables display with admirable clearness the great diversity of ages in the several grades and establish the fact of retardation by unmistakable evidence. Instructive as are the facts stated and the conclusions drawn from this valuable statistical material, they by no means exhaust its possibilities. It is the purpose of the present article to inquire what further facts can be elicited from this material, to quarry out if possible additional information which may be embedded in it. To do this will involve a scrutiny of the tables, not only in their general outlines, but in their individual parts. Such a scrutiny may also serve to illustrate the maxim that in a statistical study every figure should be challenged?not for its accuracy, but for the purposes of interpretation. The tables compiled by Dr Cornman show a general resemblance to one another, and much difference in detail. It is our purpose to call attention to both resemblance and dissimilarity. In the first place, it may be pointed out that the distribution of the ages among the different grades shows greater variations than in the several school systems as aggregates.
If we direct our attention to the total number of children of each age, as reported in the tables, we note a marked similarity in tlie series for the different cities. In each, the numbers between the ages of 8 and 12 are very nearly identical, growing smaller, however, as the age of 12 is reached. But in every case the numbers fall off somewhat at the age of 13 years, and much more obviously at 14. The following figures are drawn from the tables:
Children in School at Certain Ages. Av. Age Camden Kansas City Boston Philadelphia New York 10-12 1,401 3,228 7,G71 18,162 62,360 13 1,102 2,812 6,575 15,150 49,840 14 656 1,868 3,959 7,913 29,450 15 322 904 1,524 2,856 10,587 Av. Age ‘ 10-12 100 100 100 100 100 13 79 87 86 83 80 14 48 55 52 44 47 15 23 28 20 15 17
This table shows that the falling off in school begins at 13, and that at 14 only about one-half of those who are in school at 12 are still present. At 14 very few have entered the high school; at 15 this number is not inconsiderable, and we should have to add high School figures in order to determine how many had dropped out of school. From these facts we can draw the conclusions that dropping out of school depends more upon age than upon the degree of advancement in school studies. This is quite as true of Boston, where, according to Dr Cornman’s conclusions, retardation is least prevalent, as it is of Camden, where it appears to be most frequent.
This fact is of cardinal importance in the study of retardation. It dispels at once the conclusion, not infrequently drawn, that retardation increases the number of children in school. One can readily see how retardation would swell the number of school children, if all were obliged to remain in school until their courses were finished. But this is not the case. After the end of the compulsory school age, generally the fourteenth birthday, they can leave when they please, whatever their stage of advancement. They in fact do so in large numbers, and as the falling off at 13 years testifies, some of them anticipate the fourteenth birthday. Thus Dr Cornman says,* “Boston is now able to make the proud boast that she has a seat in school for every child able to attend. This condition may be due in part to the smaller percentage of retardation. Were the stream of children through the grades less rapid, perhaps she would have her thousands or tens of thousands upon part time, while empty benches yawned for occupancy in the highest grades. Damming the stream of children passing through the grades of our schools subverts the purposes of the public educational system and causes a wasteful expenditure of the public funds.” And again, “The child that takes ten years to complete an eight-year course, costs the state 25 per cent more than one who gets through 011 time. Complete statistics of retardation would furnish the data upon which might be made a fairly accurate computation of the cost of this delay in the grades. It is safe to assert that the total loss to the community would be found to be surprisingly large.” In these sentences we find the problem of retardation brought into relation with the problem of accommodation, and with that of cost.
]STow it is perfectly clear that if Boston must accommodate all her children till they are twelve years old inclusive, and in addition at least 85 per cent of those thirteen years old, and 50 per cent of those fourteen years old, it matters not what the stage of academic advancement may be, they will under all circumstances require the same number of seats. Looking at it from the standpoint of cost, it is equally plain that it will cost just the same to teach Boston children, whether they are well along in their studies or far behind. The argument in the final quotation seems convincing and is true for the specific case mentioned. But the specific case does not represent the average case. The child does not generally take ten years to finish an eight-year course. lie simply does not finish. Indeed there is some ground for believing that if retardation were wholly eliminated from our schools the cost would be increased. If there were no retardation and all the children in their fourteenth year were in the seventh grade, there is good reason to believe that a much larger number would be ambitious to finish the work of the common schools, and that we should have a much larger number of children in their fifteenth year in the school than we have at present.
It is not my purpose to minimize the evils of retardation, but cost is not one of them. I quite agree that it causes a “wasteful expenditure of the public funds” because it is an ineffective expenditure. The expenditure is not greater than it otherwise would be, but the results are painfully less. 60 THE PSYCHOLOGICAL CLINIC. The fact that has the greatest significance for society is not that retardation means a prolonged education measured in time for the few, but that it means a shortened education measured in distance for the many. Thus, if in Camden we find 317 pupils of eleven years of age in the third grade, the sad fact is not that they will be sixteen years old when they reach the eighth grade, but that the vast majority will never get there. They will reach their fourteenth birthday in the sixth grade, and most of them will disappear at that time. Retardation, whatever its causes, means that comparatively few children finish the work of the elementary schools. If remedial measures can be made effective, a larger number of our children will receive a full elementary education, as provided in our common schools.
The age at which children begin school is as important as the age at which they leave it, since between these two points lies the possibility of advance. We have seen that the five cities compared show substantially similar results so far as the decrease of the school population after twelve years is concerned. But do the children begin equally early in the several cities ? The answer of the figures is as follows:
City ? years g years 7 years 8 years and under Camden 611 1,270 1,556 1,434 Kansas City 3,502 ‘ 3,109 3,104 Boston 2,835 7,500 8,284 8,764 Philadelphia 8,611 16,583 17,769 New York 2,446 36,797 53,789 61,367 Here, as elsewhere in this study, information not directly derived from Dr Cornman’s tables has been obtained from the original reports of the respective cities. The addition of the group “five years and under,” when it appears, divides the group “six years and under” in Dr Cornman’s tables. In Kansas City and Philadelphia no information is available as to the children under six years of age, if there are any who have been included in the tables. The figures apparently show two types; the first three cities in the table above showing but a slight increase of the children seven and eight years of age over those of six years of age, while in New York and Philadelphia these ages are much more numerous than the six year old children.
An examination of the tables in Dr Cornman’s paper will show that the figures for Camden appear to be reported in September and those for Kansas City in June. As a matter of fact, the two tables are identical, and neither is what it is labeled. In both cities the official school reports show that the figures relate to all children admitted to the schools during the year ending June, 1906. In both cases the ages are those recorded at the time of admission. As by far the greater number entered in September, the figures for Camden are very nearly, though not exactly, what they purport to be, i.e. the ages in September. The basis of enumeration, for Kansas City and for Camden, is identical, but it is quite distinct from that used in the other cities. In Boston, Philadelphia, and New York, the figures are taken in June, and hence the children are recorded at a time when they were ten months older than those of Kansas City or Camden. Children, who in September were five and six years old, are now mostly six and seven. In Philadelphia, the admission of children to the grades when less than six years of age is infrequent, and as the figures are given in June, it is not improbable that those who entered the schools in September at six years of age were nearly all seven when their ages were recorded. The six year old children present in June would represent to a large extent those who came in with the second term in January. In New York, admission at five years of age is more frequent, hence we find in June a larger proportion of six year old children present in the schools of that city than in Philadelphia. In Boston, however, the admission of five year old children is very frequent. There wTere as many as 2,885 in June, 1906, and an even larger number of six year old children, despite the fact that the ages were recorded in June, showing the presence earlier in the year of a very much larger number of these young children. This explajins why Boston appears to conform to the type of Camden and Kansas City, although the children were recorded at the end of the school year instead of predominantly at its beginning.
This examination of the first factor in the tables, that of age, has shown that the several tables, though similar in form, are quite distinct in substance. They are not so directly comparable as they would appear. If the aggregates of the age columns are affected by these differences, the details must also be affected, and it becomes necessary to inquire whether the conclusions drawn by Dr Comman in regard to ages within the grades must not be modified, in part at least.
Before doing so, it will be well to inquire whether the tables presented are comparable in every respect as to grades. The time of the enumeration is not without influence upon this factor also, but there are other and less subtle differences revealed by a careful examination of the tables. If we compare tlie number of pupils in each grade in the several cities, we note that in Camden, Kansas City, and Boston, the first grade is the most numerous, while in Philadelphia and New York the second grade is the largest. Indeed, in Philadelphia the second grade contains more than twice the number of children found in the first, and in New York also the second grade is very much the larger. A reference to the actual figures shows:
Grade Boston Philadelphia New York June, 1906 June, 1907 June, 1908 First Grade 13,669 13,353 57,893’ Second Grade 10,276 29,929 81,780 Obviously the Philadelphia and New York figures do not represent a normal grade distribution, but this incongruity is not explained by the figures themselves. In both cases the school authorities count the children after the June promotion. To be specific, the first grade pupils reported do not comprise all of those who have been in the grade during the year, but only those first grade pupils of the school year 1906-07 who will begin the year 1907-08 in the same grade. In other words, the grade distribution reported in June is that with which the next school year would begin in September, provided no pupils dropped out during the summer and no new pupils were admitted in the fall.
Our five tables, apparently so similar, are in fact very diverse, and this diversity apparently affects the numerical relations which we arc studying. Let us therefore examine closely just what each table represents, and how their differences affect their comparability in detail. Our five tables represent at least three distinct methods of enumerating ages and grades: 1. All the children entering school during the year, grades and ages being reported as of the time of entrance (Camden find Kansas City).
2. All the children in” school at the end of the year, the ages reported being those at the end of the year and the grades those of the last half of the year (Boston).
3. All the children in school at the end of the year, the ages reported being those at the end of the year and the grades those of the next year (Philadelphia and New York).
In general it may be said that the tables differ widely on account of this absence of consensus of opinion as to the precise meaning of such simple phrases as the age of the child and the grade in which the child is found.
The first method of enumeration, which is employed in Camden and Kansas City, and which counts all the children admitted to school in the grades in which tlicy are admitted, has certain advantages and disadvantages. Its advantages are found in a more accurate determination of ages at the time the record is made. Its disadvantage lies in the fact that while the bulk of the record is made in September, it is spread over the entire year. The age record can be made in the school books by asking the child its age, or the date of its birth or both. Boston, Philadelphia and New York record the date of birth and can thus compile ages accurately at any time. This is always to be commended, but is by no means universal in the school records of American cities. If a child is recorded in September, 1907, as having been born in January, 1900, we know that in March, 1908, it belongs in the eight year old class. If, however, it is recorded in September, 1907, as being seven years of age, we do not know in March, 1908, whether it is still seven or whether it has reached eight. If the record, as is frequently the case, is a double one,? Born, January, 1900, age seven years?accuracy is possible at any later date, but unless special pains are taken to secure such accuracy, the question as to the age of the child is likely to elicit the answer that it is a seven year old child, whether the question be put in September, 1907, or in March, 1908. The teacher is apt to report the recorded age without calculating the actual age. When, therefore, the ages are reported as of the date of June or any date later than the registration date of the pupils, there is always some doubt as to their accuracy. In the matter of ages, inaccuracies are notorious. Under either system there may be errors in the original entries. The less care taken in making such records, the more frequent such errors will be. We are not dealing here with such original errors, but with secondary errors arising from mistakes or carelessness in transcribing the facts from school registers.
These considerations lead us to believe that if the ages of children entered in September Were reported on October 1st as of the date of registration, they would in a great majority of cities represent more accurately the school records, than if they were reported in June as of that date.
But if, under the system pursued in Camden and Kansas City, the bulk of the age figures refer to a specific date- in the school year, namely, in September, many of them do not. The school register of September is continually growing through the year. In this system of records both age and grade are reported at the beginning of the child’s school life for the given year. In the great majority of cases this will be in September, but in quite a number of cases it is later. In Camden, during the year in question, the total registration in elementary schools was 12,801, but the average registration was only 9,864. From the manuscript records of his office, Superintendent Bryan has furnished me with the following interesting record of the growth of the total enrolment, showing the exact number of pupils enrolled from the beginning of the year to the end of each month during the school year 1905-1906, from which the monthly accessions can be derived:
End of Total Enrolment Accessions September, 1905 11,117 11,117 October, 1905 11,627 510 November, 1905 11,822 195 December, 1905 11,952 130 January, 1906 12,122 170 February, 1906 12,400 278 March, 1906 12,508 108 April, 1906 12,696 188 May, 1906 12,719 23 June, 1906 12,801 82
It appears that of the pupils registered during the year, 86.8 per cent, or very nearly seven-eighths, were in the schools in September and were recorded that month as to their ages. Accessions were numerous in October and over 90 per cent of all enrolled were recorded before November 1st. There is, however, a scattering contingent whose ages are recorded at later dates down to the close of the year in June. Evidence that similar conditions existed in Kansas City is found in the fact that among 28,564 children entered in the schools, there were 6,076 who attended less than half the school year, many of them doubtless because they entered after the year’s work was well under way. What has been explained at some length in regard to age distribution is true also of grade distribution. The facts were reported when the child entered school, and subsequent changes were not noted. It may be observed that in the case of a school system having half-yearly or more frequent promotions during the year, this method offers considerable difficulty. Who shall report the child?the teacher who has him under her care at the end of the year, or the one who began with him? Obviously, the more definite system adopted by tlie city of Boston avoids tliese difficulties. A second method of stating the facts prevails in Boston. Here the statistics are prepared semi-annually on January 31st and June 30th, and the grades reported are those in which the pupils have been during the previous five months. The ages reported are those of the date of enumeration. In Dr Cornman’s paper the analysis is based upon the June figures of the year 1906*
We have here two important elements of differences from the first method. It is fair to presume that some of the pupils who in September were over-age have dropped out. This would have a tendency to reduce the number of over-age pupils were the ages reported those of September. But all the pupils recorded are nine or ten months older than they were in September, and this would increase the number of over-age pupils. Fortunately we have some means of estimating the force of these conflicting tendencies. In table VI Dr Cornman prints percentages for Boston in January, 1906, as well as those of June, 1906, though he does not comment upon these added figures.
In June, 1906, the percentage of over-age pupils in Boston as a whole was 21.6 per cent, but in January it was only 15.4 per cent. In the first grade there were in June 9.7 per cent pupils over-age, but in January only 5.7 per cent, and corresponding differences are found in all the grades. In short, the January pupils are five months younger than the same pupils in June. In September the pupils are again five months younger.
This is strikingly shown in the figures for the youngest children in school. Thus we find, comparing January with June, 1906:
Ages Number of Children January June 4 years of age 67 20 5 years of age 5,470 2,815 G years of age 8,242 7,500
Since most of the five year old children were nearing the age of six in January, they decrease more rapidly in number than the other age classes in the following five months. There are no comprehensive figures for September, 1906, but a special investi*It is to be noted that in the year 1906 there were nine grades, the ninth grade containing 4,408 pupils, of whom 795 were over-age. As the percentage of over-age pupils tends to diminish in the upper grades the inclusion of the ninth grade would have slightly reduced, in the decimal, the percentage of over-age pupils. Dr Cornman confines his table to the first eight grades.
gation of pupils ten years of age and over in the three primary grades, enables us to establish the following comparison: Pupils Ten Years of Age and Over in 1900. January 31 June 30 Sept. 12 First Grade 74 81 38 Second Grade 246 429 104 Third Grade 1,099 1,789 477
The figures are arranged chronologically. The last, which represent the beginning of the school year 190G-07, are very different from those of June, representing the end of the school yeark 1905-190G. They show us a much lower percentage of retardation, and were figures available for Boston on the same plan as in Camden and Kansas City the contrast with the two latter would have been even more striking than that reported by Dr Cornman. Among the five cities compared, Boston has a unique place. Elsewhere one-half of the children are above normal age, but in Boston this proportion appears to be only one-eighfh. The comparison is highly stimulating to those who are seeking to diminish the proportion of retardation in our schools. It would appear that these gratifying results are due to the habit which prevails in Boston of sending many children to school at the age of five. It is a well nigh universal practice for children to be in the first grade soon after the sixth birthday. This early start in school life permits a child to be held back a year or so, in many cases even two, without bringing it into the over-age class. The tables for ISTew York and Philadelphia present a third method of statement much less satisfactory than the preceding, but .one which is capable, in the absence of other data, of throwing a great deal of light upon the conditions in those schools. The moment chosen for the enumeration is the least representative of general school conditions of the year. It is the moment of closing school in June when a large number of pupils are credited to higher classes, which they will not enter until school commences in September.
Compared with Boston, the ages are the same?those at the end of the year?but the grades to which the children are credited are in many cases more advanced. We have approximately a September grade distribution and June ages. Let us see how. close we approach to a September grade distribution. It would be an exact one if the children all came back in September and no new pupils entered. Of course many pupils will not return in September. Of those who fall out some will be in the upper grades. Of these it is not unlikely that the over-age pupils will be the most numerous. The tendency to leave school after fourPHILADELPHIA. Estimated Figures for September, 1907, for Distribution of Pupils in the Grades by Ages. teen is very strong. There is no doubt that it is stronger among the relatively unsuccessful than among those “who have reached a place in the school corresponding to their ago.
On the other hand, many new pupils will appear in September. The ranks of the first, grade depleted by the June promotions will again fill np, and this grade will be the most numerous in the entire system. But accessions to higher grades will be relatively few. They will offset the losses, but it is not to be surmised that they will change the general age distribution in the grades from the second to the sixth inclusive. If therefore our figures represent approximately a September grade distribution with the exception of the first grade, we must add to that grade in order to make a closer approach to the real facts. Among the added pupils it is safe to assume that only a very small proportion will be nine years of age, all the rest being younger. The ages of our table, however, will still be those of June, just which day in June is not apparent, but probably not as late as the thirtieth of the month. !Nor will the schools open precisely on September 1st. There is an interval of somewhat more than two months: to he on the safe side let ns assume that it is three months. Now of all the children of eight years and less than nine, it is safe to assume that three months later onefourth of them will have passed their ninth birthday. Raising one-fourth of the eight year children to nine years and so on throughout the entire table, the Philadelphia figures for June, 1907 (Cornman’s table, Vol. I, p. 249), are transformed into the table for September, 1907, which is given on the preceding page. The full import of this table can only be seen by comparing its main results with those published by Dr Cornman, as follows: Percentage of Enrolment Above Normal Age.
Camden Philadelphia, Pa. Grades mainly June, ‘07 September, ‘07 September, ‘06 Estimated 8 45.2 23.9 31.6 7 61.0 31.1 39.3 6 59.1 41.5 48.6 5 63.7 45.1 51.7 4 63.6 43.0 49.8 3 55.5 39.3 46.1 2 44.4 28.3 34.2 1 26.8 36.5 37.9 Totals 47.5 37.1 42.5 In general it will be seen that the estimated figures for September are from five to six points higher than those of June and that they thus approach the figures for Camden, N”. J., and Kansas City. If the first grade figures (and therefore the total figures) contain a large conjectural element, it should be remembered that in the upper grades this is much less. But the broad fact remains that if the grades had been stated in the manner employed in Boston the percentage of retardation would have appeared larger than had the Camden method been employed. Proof of the latter statement is given by Dr Cornman in table VI. In this table he gives three percentages for New York, showing a total retardation in June, 1904, of 39.0 per cent, in June, 1905, 32.0 per cent,* and in June, 1906, 30.0 per cent. Dr Cornman suggests that improvement has followed calling attention to the problem. The ~New York reports, however, show in ?From N. Y. City Report, erroneously printed as 30.1 in Dr Cornman’s paper.
1904 the pupils before promotions, as in Boston, but in the following year, for some unexplained reason, the method was changed and the grades were noted after the promotions had taken place. The facts are presumably about the same in each case, but the former method gives a percentage nine points higher than the second. We have examined the figures thus carefully in order to eliminate as far as possible the discrepancies due to differences of method. Bringing all the figures as nearly as possible to the total enrolment basis employed in Camden and Kansas City, we have the following result: Percentage of Retardation Different Methods Total Enrolment Camden 47.5 47.5 Kansas City 49.6 49.6 Boston 21.6 12.5* Philadelphia 37.1 42.5* New York 30.0 35.0* *Estiinated. A possible cause of variation among cities may lie in the extent to which ungraded classes were represented in the city schools systems. If all the backward children were put in such classes, retardation would disappear from the regular grades. Only a beginning has been made with such classes, and their influence upon the general result is not very great. Thus at the time of the enumerations cited we find the following: In Regular Grades In Special Classes Boston 71,377 2,680 New York 500,076 19,679 Information for Philadelphia in June, 1907, is not available. On December 31, 1906, there were 833 children in special classes, about two-thirds of whom were in classes for truants and incorrigibles and about one-tliird in classes for backward pupils. The aggregate elementary school population at that date was 155,763.
If all the special classes were added to the backward children in New York in 1906, the total proportion of backwardness would be something over 32 per cent instead of 30 per cent. It is to be noted that the figures reported for 1905 are 32 per cent and that in that year there were no special classes.
It may seem to some that this discussion of statistical methods has been unduly prolonged, but only by such discussion can the true meaning of the figures be ascertained. It is certainly most unsatisfactory to have to deal with divergent figures which can be made strictly comparable only by introducing conjectural elements. In view of the increasing interest in the problems involved, it would certainly bo gratifying for all concerned were school men to follow the same method.
Which method is to be preferred? Among the three whicK have been discussed, that of Boston has the merit of the greatest simplicity. Its disadvantage is that it records facts at the end of the year. If a particular time is to be selected to describe a given grade, it is clearly not the end of a year which is most appropriate. The proper expression for the population of Boston in 1907 would be neither that of January 1, 1907, nor that of December 31st, but plainly that of June 30th, in the middle of the calendar year. Attention has already been called to the difficulty of transcribing accurately at the end of the year the facts as to age as they are commonly reported in American school registers.
Again the moment after the June promotions selected in Philadelphia and New York is distinctly not typical of the year. ISTo other date in the whole school year is less characteristic than this. The total enrolment recorded in Camden and Kansas City does not, as it is presumably intended to do, represent the year as a whole, for it records ages and grades at a wide variety of dates. Recognizing the superior principle involved in the Boston method, it seems to the writer that the faults of that method would be obviated by applying it on October 1st. By that date the regular fall registration is complete. At that period of the year a transcription of the age record will be at least as accurate as the original entries, whatever be the method of making those entries. Another consideration of moment is that the school population reaches its maximum about the first of October. As the year progresses some of the children drop out and while the places of some are filled by new recruits, the accessions do not counterbalance the losses, and at the end of the year the school population is at its lowest ebb. The ages of the pupils who drop out and the effect of such losses upon the age distribution is something which, so far as I have been able to ascertain, has not been determined. It is safe to infer that the ages of those dropping out are on the average higher than those of the new accessions. The accessions always contain a large proportion of beginners. On the other hand, the losses contain at least a considerable number of those who leave school never to return, again,?who are, in short, over-age pupils. This does not, however, account for all the losses. It does not explain why the first and second grades are frequently less well filled at the end of the year than at the beginning. In the years of optional school attendance before eight years and after fourteen, there is undoubtedly a greater movement in the school population than in the ages eight to thirteen inclusive. Of course losses of pupils over fourteen years of age are not. compensated by accessions, while losses of those under eight are largely offset by additions to the register.
The advantage of recording the pupils at the beginning rather than the end of the school year lies in the fact that on a yearly promotion basis all the pupils are beginning for the year the grade to which they are credited, while at the end of the year they are finishing it. “When a half yearly promotion system prevails, the difference between the two methods is less. On the whole, the children have a more permanent relation at the beginning of the year to the grade in which they are reported than at the end. In a discussion which is perhaps more theoretical than I could have wished, it may seem out of place to inject purely practical considerations. 33ut nevertheless I would like to point out from practical experience in administering a school system the great advantage of making such detailed statistical records early in the year, so that they can be accurately summarized and analyzed in the scliol report.*
Too much emphasis cannot be laid upon simplicity and uniformity of method. If an agreement could be reached upon this topic, it would greatly promote our knowledge of the subject. Such criticisms of method as have been the chief theme of this paper would be avoided and we should have more time to study the facts, to probe after the causes, and to test possible remedies.
Up to the present time the statistical study of retardation has largely been confined to a’ discussion of the facts. It is perhaps too early to seek to measure causes. Those who have given ?The writer’s practice in the schools of Porto Rico wliere the facts are recorded on the first of March was not in accord with the suggested date. The reasons for selecting a later date in Porto Rico was that \ e had to do there with a mixed system of town and country schools, that tlie schools, and especially those of the rural districts, were often la e in getting under way and that March represented the maximum eniolment of the schools of the island as a whole. The maximum for town schools was found there to be in October. In a purely city system this fact should determine the time of making the record.
the matter attention have indicated a certain number of causes without trying to estimate their relative force. In this connection the article of Dr Cornman, on which the present study is based, is highly suggestive, and I am tempted to add a few considerations 011 this part of the subject. It might be noted that in his introductory statements, Dr. Cornman unconsciously falls into the error of explaining retardation by the fact that pupils are not promoted regularly. Is it not better to insist tljat “retardation” should always be used to express a fact and is not an explanation ? “Retardation” means a lack of correspondence between grade and age. Unless it is so used, late entrance into school cannot be given as an explanation. A boy who entered school at ten years of age and advanced regularly would be “retarded” in the technical sense, though his progress would be normal and he would not be held back by failure to secure promotion.
Later on Dr Cornman says, “A certain proportion of the retardation is due to the fact that pupils are already over age on entering school,” and cites the proposal to make the age of compulsory attendance seven instead of eight years. While the advantage of an early start in school is undeniable, as the experience of Boston amply proves, it is doubtful whether so slight a change in the law as that noted, would have any appreciable effect. It is to be noted that in Philadelphia and New York, at the very end of the year, the number of eight year old children is not much larger than that of seven year olds. New Jersey, as well as Massachusetts, makes school attendance compulsory at seven, yet Camden and Boston show vastly different results.
There is another factor bearing upon the retardation of pupils which has thus far received scant attention. It is the irregularity of school membership. The school population contains many fluctuating elements. One of the fundamental conditions of regular advance is membership throughout the year in the same school. It is surprising to those who have never examined the subject closely to find how large a proportion of all the children enrolled in the schools are there for periods less than the school year. Some, as we have already seen, come in late. Many others drop out in the course of the year. In so far as the latter are younger than fourteen years, it is quite likely that they will return to school after an interval. Such a loss of time, whether caused by illness or other reasons, puts the pupil behind in his work and swells the number of over-age pupils. While we have no exact measurement of the force of this factor, there are some interesting indications that it is one of considerable moment.
In Camden in 1905-06 the total register grew from 11,117 at the end of September to 12,801 at the end of June. There were therefore 1,694 accessions during the year. But at the same time there were children falling out during the entire year, and in the month of June there were only 9,729 children on the rolls. This shows us that of the 12,801 children on the rolls only 8,035 were registered continuously throughout the year. In addition to these, as we have already seen, 1,604 came in after the year had opened, while 3,072 left during the course of the year. As a matter of fact, although the statistics refer to 12,801 children, at no period of the year were there so many children in school. From manuscript records Superintendent Bryan has furnished me the folloiwng interesting figures:
1905 Total Roll for Month 1906 Total Roll for Month September 11,118 January 11,072 October 11,584 February 11,221 November 11,393 March 11,073 December 10,935 April …… 10,934 May 10,548 June 9,729
Somewhat less complete but equally suggestive are the figures available for Kansas City. In the year 1905-06, there were 28,564 registered, though the average daily membership of the schools was only 22,720. The following table is given of the length of attendance:
Attended 200 days 1,367 Attended 180 days and less than 200 days 12,276 Attended 160 days and less than 180 days 4,000 Attended 140 days and less than 160 days 2,059 Attended 120 days and less than 140 days 1,437 Attended 100 days and less than 120 days 1,349 Attended 80 days and less than 100 days 1,273 Attended 60 days and less than 80 days 1,130 Attended 40 days and less than 60 days 1,123 Attended 20 days and less than 40 days 1,314 Attended less than 20 days 1,236 28,564
Now it is to be observed that attendance and school membership are not identical. It is quite possible that the 17,643 children who attended 1G0 days or more were continually on the roll, and also that they received substantially the benefits of a year’s teaching despite occasional absences. But when the attendance is much less, this is due to the fact that some children were on the rolls of the school but a short time. If, for instance, we can expect 90 per cent of the pupils to advance in their grades, this could only be anticipated of those who had enjoyed substantially the year’s work. Those who entered for shorter periods must go to swell the number of over-age pupils.
The facts to which attention is here called are found in all our cities. Boston, for instance, had in 1905-06 a total enrolment of 89,130, but on June 30, 1906, the number in the schools was 78,465. New York and Philadelphia show a similar state of affairs. It is not for the purpose of demonstrating its existence that the figures have been cited, but to bring out the strong bearing of this irregular membership on the problem of retardation. The subject of retardation in our elementary schools has too recently gome into the foreground of public interest to essay a comprehensive plan for the study of the facts and their causes. The analysis of the figures presented from five different cities shows that, whatever form of stating the facts may ultimately win general approval, there can be no doubt that a uniform method is highly desirable. As to what that method should be, it is the belief of the writer that the enumeration of the children actually in school at a given date soon after September 1st presents decided advantages, especially from an administrative standpoint, over other methods which have been followed. In the matter of statistically measuring the causes of retardation, we are entirely in the experimental stage. It has therefore seemed worth while to call attention to the influence of irregular school membership, in order that this important factor may not remain unnoticed by those who are giving their time and thought to the problem in all its aspects. s
Disclaimer
The historical material in this project falls into one of three categories for clearances and permissions:
Material currently under copyright, made available with a Creative Commons license chosen by the publisher.
Material that is in the public domain
Material identified by the Welcome Trust as an Orphan Work, made available with a Creative Commons Attribution-NonCommercial 4.0 International License.
While we are in the process of adding metadata to the articles, please check the article at its original source for specific copyrights.