The Phenomenon of Scattering in the Binet-Simon Scale
By J. E. Wallace Wallin, Ph.D., Psycho-Educational Clinic and Special Schools, Board of Education, St. Louis.
It has been said that feebleminded children are more uneven in their intellectual development than normal children. According to Binet and Simon1 “a defective child does not resemble in any way a normal one whose development has been retarded or arrested. He is inferior, not in degree, but in kind. The retardation of his development has not been uniform. Obstructed in one direction his development has progressed in others. … So far as certain faculties are concerned, he remains at the level of a younger child; but in respect to others, he is on a level with normal children of his own age. An unequal and imperfect development is consequently his Specific characteristic. These inequalities of development may Vary to any degree in different subjects. They always produce a want of equilibrium, and this want is the differentiating attribute of the defective.” The difference between the feebleminded and the normal is qualitative rather than quantitative.
We assume it is on the basis of these sweeping generalizations, which, we are led to conclude, represent merely the personal opinion of the authors as no supporting facts are adduced, that the conclusion has been drawn that the feebleminded “scatter” more in the BinetSimon scale than do normal children, i. e. they fail on more tests m lower age levels while passing more tests in various age levels above the basal age than do the normals. This alleged fact has almost been given the significance of a pathognomonic sign by which we are able to identify those who are “potentially feebleminded,” and by which we are thus enabled to make an early prognosis of feeblemindedness in the case of doubtful children who scatter” considerably, but who cannot be diagnosed as feebleminded on the basis of the degree of mental deficiency which they evince. Thus we find Doll2 affirming that “In general a normal child develops intellectually more consisently than a potential defective. The component mental processes which determine the intellectual capacity develop uniformly in the former and not so 1 Mentally Defective Children, 1914, p. 13.
Preliminary Note on tbe Diagnosis of Potential Feeblemindedness, Training School Bulletin, 1916, May. uniformly in mental defectives. In the Binet tests the typical normal child has a basal year not more than one year below his chronological age and passes but a few tests beyond his chronological age. The potential feebleminded, on the other hand, has the basal year more than one year below the chronological age or at least seriously below the total mental age rating and may have more than one basal year, that is to say, he ‘scatters,’ failing in tests one would expect a normal child of that age to pass, and succeeding in others not expected. Thus a normal child ‘testing’ 8 may have a basal year of 7 and pass 3 tests at 8 and 2 at 9, whereas the potential feebleminded ‘testing’ 8 may have a basal year of 5, pass 4 tests at 6, 4 at 7, 3 at 8, 2 at 9, and 2 at 10. The normal child thus tests 7.5 and the potential feebleminded 5.15, the total rating in each case being1 8. This brings us to a second premise, namely, that although the individual Binet tests at any age are of equal difficulty for normal children, they are of somewhat unequal difficulty for mental defectives.”
It is evident that if unevenness in a child’s mental development, as indicated by scattering on the Binet-Simon scale, is an indication of potential or actual feeblemindedness, it is worth while determining whether feebleminded children do, in point of fact, “scatter” more than normal children. We do not know that any one has thus far presented any experimental data in support of the correctness of this hypothesis. Doll’s conclusions are admittedly tentative: “It is at present only a result of observation that such differences do exist and experiment has too often demonstrated that observation is mistaken.” We have ourselves already called attention to the extensive scattering on the Binet-Simon scale by epileptics (which we noted as early as 1910) and by insane subjects.1 From our data which have thus far appeared in print it has not been possible to determine the extent of the excess of scattering found among the epileptics and the insane as compared with other classes of persons. During the last eight years we have occasionally noticed that some feebleminded subjects scatter very widely on the BinetSimon scale, while other feebleminded subjects scatter very little. We, however, deemed it inadvisable to attempt to compile definite figures until we had a considerable accumulation of data available. We now have records for thousands of subjects whom we have personally examined and diagnosed (with the aid of cooperating physicians), but, owing to the lack of time to work up all our records Experimental Studies of Mental Defectives, 1912, p. 22. Problems of Subnormality, 1917. p. 161f. The Individual Tests in the Binet-Simon Scale, The Psycho1.. Clinic, 1917, p. 79. Wide Range versus Narrow Range Binet-Simon Testing, Journal of Delinquency, 1917, 315f.
(and the unfortunate circumstance that nearly a thousand of our records are not now accessible to us), we are obliged to limit this study to 840 consecutive1 cases examined in the Psycho-Educational Clinic conducted by the Board of Education in the City of St. Louis. These cases were diagnosed as follows: normal, 5.5 per cent; retarded, 7.9 per cent; backward, 38 per cent; borderline, 12.5 per cent; deferred, 6.4 per cent; and feebleminded, 29.4 per cent (morons, 17.1 per cent and imbeciles 12.2 per cent). The average chronological age of the boys was 11.10 years; of the girls, 10.87; and of both sexes, 11.07.
Results. Table I shows the number of advance points which were earned by the pupils in the different classifications. The results differ somewhat according as we follow the 1908 scale or 1911 (Vineland) scale. Turning to the 1908 averages for the two sexes, we find that the smallest number of advance points were won by the pupils designated as normal, followed by the borderline and backward, who earned the same number of points, while the highest number of points were earned by the “deferred” cases (i. e. the diagnosis was deferred because the classification was uncertain) and the morons. The same number of points were earned by the retarded and the imbecile children. The greatest difference in the number of advance points earned is between the normal and the deferred, amounting to .58 of a year. Exclusive of these two categories the greatest difference amounts to one-fifth of a year, or only one point. The difference between the four of the seven .categories which show the smallest differences varies from 0 to .04 of a year, an insignificant amount.
The smallest number of advance points among the boys, 1908 scale, was earned by the normal cases, followed by the retarded and backward whose averages are the same; and among the girls, by the normal, followed by the borderline. The greatest number of advance points among the boys were earned by the deferred, followed by the morons (the borderline and the imbeciles earned the same number of points); and among the girls, by the deferred, followed by the retarded and morons, who earned equally many points. The greatest difference among the boys amounted to .46 of a year, and among the girls .86 of a year, in both cases between the normal and the deferred. Exclusive of the two extreme categories, the greatest difference amounts to .18 of a year among the boys * The exceptions to this statement have been noted in Wide Range versus Narrow Range Binet-Simon Testing,” Journal of Delinquency, 1917, p. 315f, which should be consulted for other facts which we do not repeat here. The subjects here considered are those included in11 Group I’ in that article. and .34 of a year among the girls. The normals earned the fewest and the deferred the greatest number of advance credits among both the boys and the girls. The difference between the categories seems to be slightly larger among the girls than among the boys. Table I.?Advance Points Earned According to the 1908 and 1911 Scales.
Normal Retarded.. Backward.. Deferred… Borderline. Morons Imbeciles.. Boys No. 37 53 246 35 81 93 59 1908 Points 4.1 4.5 4.5 6.4 4.8 5.4 4.8 1911 Points 8.0 7.0 7.3 7.1 6.9 6.5 6.8 G:rl? No. 1908 Points 2.9 6.5 6.0 7.2 4.8 6.5 5.3 1911 Points 7.3 6.0 7.0 6.8 6.6 6.2 6.2 Both No. 47 67 320 54 105 144 103 1908 Points 3.8 5.0 4.8 6.7 4.8 5.8 5.0 1911 Points 7.8 6.8 7.2 7.0 6.8 6.3 6.5 1908 Years .76 1.00 1.34 1.16 1.00 1911 Years 1.56 1.36 1.44 1.40 1.36 1.26 1.30
To convert the advance credits in points to advance credits in years multiply the points by .2. This has been done in the last two columns for the two sexes.
The figures for the points represent averages.
Turning to the 1911 scale, averages for both sexes, the smallest number of advance points were earned by the morons, followed by the imbeciles, while the greatest number were earned by the normal, followed by the backward. The same number of points were earned by the retarded and borderline. The greatest difference in the number of advance points earned is between the normal and borderline, amounting to .30 of a year. Exclusive of these two categories the greatest difference amounts to only .14 of a year, while the difference between those four of the seven categories which show the smallest difference varies from 0 to .10?differences which, in either case, are rather negligible.
For the boys, 1911 scale, the smallest scattering is among the morons, followed by the imbeciles; and for the girls, among the retarded, followed by the morons and imbeciles, whose averages are the same. The greatest scattering for both the boys and the girls is among the normals, followed by the backward. (The scattering is the same among the girl morons and imbeciles.) The result for the two sexes are in complete agreement with respect to the groups which scatter most, but in less close agreement with respect to the groups which scatter least. The extreme difference among the boys amounts to .30 of a year, between the normal and the morons; and among the girls, .26, between the normal and the retarded. Exclusive of the two extreme categories, the greatest difference amounts to only .10 of a year for the boys and .16 of a year for the girls. With the 1911 scale the difference between the boys and the girls is not significant, while with the 1908 scale the variation between the categories was slightly greater for the girls. The difference between the different categories is greater with the 1908 than with the 1911 scale for both the boys and the girls.
It is apparent that the two scales yield discrepant results so far as the mean tendencies are concerned. According to the 1908 scale it is the deferred and the morons, but according to the 1911 scale the normal and the backward subjects who scatter most, while according to the 1908 scale, it is the normal, the borderline, and the backward, and according to the 1911 scale the morons and the imbeciles, who scatter least. The difference between the extreme categories is greater with the 1908 scale than with the 1911 scale, but the differences are not’very pronounced with either scale.
In order to lessen the number of categories, which may bring out more clearly the alleged tendency of the feebleminded or subnormal to scatter more extensively than the normal, we have reclassified the subjects in table II into three groups. The “normal” group comprises the backward and the retarded, who are certainly not feebleminded, as well as those who were technically classified as “normal” in table I. The subnormal group includes the borderline, deferred, morons, and imbeciles, all clearly below normal, while the feebleminded group is limited to those who have been classified as morons and imbeciles.
Based on the averages for the two sexes the greatest scattering occurs among the normals and the least among the feebleminded, when using the 1911 scale. But the difference amounts to only ?16 of a year. The difference between the normal and the subnormal categories is greater (.12 of a year) than the difference between the subnormal and the feebleminded (.04). On the other hand, when using the 1908 scale the scattering is least among the normals, while it is the same among the subnormal and the feebleminded. The extreme difference, again, amounts to only .16 of a year. With the 1908 scale the least scattering for both the boys and the girls occurs among the normals. The scattering is the same among the subnormal and feebleminded for the boys, but it is slightly greater among the subnormal than among the feebleminded for the girls. The greatest difference among the boys amounts to only .14 of a year, and among the girls to only .04 of a year.
Table II.?Advance Points earned according to the 1908 and 1911 Scales. Bots Normal (including ‘backward,” “retarded” and “normal”) Subnormal (including “morons,” “imbeciles,” borderline” and “deferred”) Feebleminded (including “morons’’ and ” imbeciles “) No. 1908 Pts. Yrs. 1.04 1.04 1911 Pts. Yra. 7.3 6.6 1.46 1.36 1.32 Girls No. 138 1908 Pts. Yr3. 1.16 1.18 1.20 1911 Pts. Yrs. 1.38 1.26 1.24 Both Sexes No. 1908 Pts. Yrs. .94 1.10 1.10 1911 Pts. Yrs 1.44 1.32 1.28
The figures for the points and years are averages. With the 1911 scale, the scattering is the smallest among the feebleminded and the greatest among the normals for both the boys and the girls. The greatest difference among both the boys and the girls amounts to only .14 of a year.
The above data are based on the averages?the average number of advance points gained by the average child in each classification. As is well known, however, the average sometimes obscures the real nature of the phenomenon under investigation. It tells us nothing respecting the frequency or distribution of the different scores or measures. It is therefore desirable to present the data in the form of a table of frequencies. Table III shows the number of subjects in each classification who earned from no advance point to 18 advance points. It is possible to determine from this table the proportion of subjects in any classification who earned one, two, three, four, or any other given number of advance points. A casual inspection of table III shows that as many as 16 advance points were scored by individual normal, backward and “deferred” subjects with the use of the 1908 scale, while with the use of the 1911 scale one backward subject earned 18 advance points, and one moron 17 advance points. The figures in this table, as arranged, however, are quite unwieldy. In order to throw the outstanding facts into bolder perspective, we give in table IV the percentage of the subjects in each classification who earned less than six advance points (i. e. 5 or less), less than eleven points (i. e. 10 or less), more than ten points {%. e. 11 or more), and from six to ten points, inclusive. The percentages have been computed only in round numbers.
We again find certain discrepancies between the 1908 and 1911 scales. According to the 1908 scale the proportion of subjects Table III.?Distribution of Advance Points earned by Subjects in Different Classifications.
ACCORDING TO THE 1908 SCALE Normal Retarded Backward Deferred Borderline Morons Imbeciles ACCORDING TO THE 1911 SCALE 0 11112 11 11 1 116171 1 213 21 1 22 10 5 15 1122 2437 3123516 18 2 20 21331435 44 4628 16 3 19 528639 19 6 25 5 2 2 8 2 10 23 9 32 4 1 5 11 7 18 8 5 13 6 5 2 7 6 1 7 35 9 44 1 4 5 14 3 17 13 7 20 7 3 1 4 3 1 4 28 7 35 1 2 3 16 3 19 13 7 20 87 7516 25 13 38 63982 10 771 9 3 3 6 6 1 7 17 11 28 5 1 6 10 2 12 11 3 10 1125 5 23 7 30 33662871 11 2134 4 12 2 14 2 22 243 12 3 1 4 13 2 15 1 1 2 2 2 13 2 2 10 1 11 1 1 1 1 14 2 2 4 4 15 3 3 3 1 4 1 1 1 1 16 112 1 1 17 18 1 1
who passed the greatest number of advance tests (i. e. more than 10 each), was greatest for the deferred subjects and the morons, and least for the imbeciles and borderline. On the other hand, the proportion of subjects who were able to pass only five advance tests or less, was greatest for the normal and backward (followed closely by the imbeciles, borderline and retarded), and least for the deferred and morons. These results are fairly consistent with those based on the averages for the 1908 scale given on p. 181.
According to the 1911 scale, the proportion of subjects, each of whom passed more than 10 tests, was greatest among the normal and backward (followed closely by the retarded), and least among the borderline and morons (followed closely by the imbeciles). On the other hand, the proportion of subjects among those who were able to pass only five advance tests or less, was greatest for the retarded and morons (followed closely by the deferred and imbeciles), and least for the normal and backward. These results, again, are fairly consistent with the results based on the averages for the 1911 scale given on pp. 182-3.
Table IV.?Percentages op Subjects earning Number of Advance Points AS INDICATED. Normal Retarded.. Backward. Deferred.. Borderline. Morona… Imbeciles.. 1908 Scam Less than 6 Points 74 61 63 38 61 47 62 than 11 Points 91 91 93 85 94 86 99 More than 10 Points 6 14 5 13 0.9 From 6 to 10 Points 1911 Scale Less than G Points Less More tlwn I than 11 Points lOPoints 21 14 15 11 5 7 9 From 6 to 10 Points
The difference between the group which showed the highest percentage who passed more than ten advance tests and the group which showed the lowest percentage was 13.1 per cent with the 1908 scale and 16 per cent with the 1911 scale.
According to the 1908 scale the proportion who passed less than 11 advance tests was greatest among the imbeciles and least among the deferred and morons, while according to the 1911 scale the proportion was greatest among the borderline and morons and Wst among the normal and backward. The proportion of those who passed from 6 to 10 advance tests according to the 1908 scale was greatest among the deferred and morons (closely followed by the imbeciles), and least among the normal, retarded and backward; while according to the 1911 scale the ratio was greatest among the borderline and imbeciles (closely followed by the normals), and least among the retarded and morons.
It is evident, therefore, that the scattering according to the 1908 scale is greatest among the morons and deferred children, some of the latter of whom will probably later prove to be feebleminded, and least among the imbeciles and the borderline, some in the later group being cases of so-called potential feeblemindedness. On the other hand, the scattering according to the 1911 scale is greatest among the normal cases (specifically the “normal,” “retarded” and “backward”) and least among the mental defectives?i. e. the “borderline,” “morons” and “imbeciles.” Most of these results harmonize with those based on the analyses of the averages, pp. 181 to 184.
If we adhere to the assumption which is usually made that the 1911 scale is more accurate than the 1908 scale (we shall consider this assumption in a separate communication) we should have to conclude that a high degree of scattering is not symptomatic of either feeblemindedness or potential feeblemindedness. Owing to the lack of higher tests in the 1908 scale, we are probably justified in rejecting the results bearing on the limited amount of scattering among the normals when using this scale, as will appear from the analysis of the mental status of the normal cases by the 1908 scale, to be made presently.
The data we have tabulated permit us to make yet another comparison. In table Y we give the average range of ages in which one or more tests were passed by the subjects in the different classification. The “range” for each subject includes all the ages between and including the basal age and the highest age in which at least one test was passed. It does not necessarily imply that one or more tests were passed in all the ages between the “basal” and the “highest” age, although this was usually the case. In the 1911 scale the “adult” age was rated for the sake of the computation as age 15, while the supposititious ages 13 and 14 were counted as two ages. In consequence of this arbitrary adjustment we count less ages than we would be entitled to do in the first case (since the adult age is assumed to be higher than Age XV, which is supplied in the scale), but more ages in the second case than actually exist in the scale when tests are passed in Age XV or the adult age. In all cases except one, however, when a plus was scored in XV, a plus was also scored in the adult age. On the other hand, the absence of tests in Ages XIII and XIV makes it impossible for subjects to pass tests in those ages who would have done so had the tests been supplied. An earlier study has shown that so uneven is mental development that children will score passes not only among the easier but also among the more difficult of higher age tests.1 We should also emphasize that the difference between the higher ages is, in general, smaller than the difference between the lower ages, while Age XV and the adult age are largely hypothetical. It is evident that the errors which arise from the above shortcomings affect particularly the older or more intelligent cases. Space limitations will restrict the comparisons to the averages for the two sexes.
Table V.?Average Number op Ages in which Tests were passed. Bots 1908 1911 Girls 1908 1911 Both Sexes 1908 1911 Normal… Retarded.. Backward. Deferred.. Borderline. Morons… Imbeciles.. 2.7 3.1 3.3 3.6 3.3 3.6 3.5 4.1 4.2 4.6 4.2 4.1 4.0 4.0 2.5 3.5 3.6 3.8 2.8 3.5 3.7 4.0 3.7 3.9 3.7 3.6 3.9 3.8 2.6 3.2 3.4 3.5 3.2 3.6 3.6 4.1 4.1 4.4 4.1 4.0 4.0 3.9
According to the 1908 scale the lowest average range of ages in which tests were passed is found among the normal subjects, and the largest range among the deferred cases, followed closely by the morons and imbeciles. The difference between the normal and the deferred amounts to 1.2 of a year, while the maximum difference between the other classifications amounts to only .4 year. We cannot conclude from these data, however, that the scattering is confined to the smallest range of ages among the normals, for an examination of the unabridged tables, which are not here reproduced, shows that a large proportion of the normal subjects graded in the higher Binet ages?by which we refer to a base of X or above?
while a merely negligible number of those in the three other classifications reached a X-year base. Forty-two per cent of the normals ‘Wide Range versus Narrow Range Binet-Simon Testing, Journal of Delinquency, 1917, p. 315ff. SCATTERING IN THE BINET-SIMON SCALE. 189 had a base of from X to XIII,1 while only 1.8 per cent (one case) of the deferred and 7.6 per cent (11 cases) of the morons reached a X-year base. The basal age for most of the deferred wap below VI. None of the morons reached X except those with a X-year base. Only one imbecile reached a VH-year base and only two a VI-year base. It is clear, therefore, that the reason the passed tests did not scatter over a wider range of ages for the normal subjects is that almost half of them had such a high base that it was impossible to give them as many higher tests as could be given to the subjects in the other groups, owing to the lack of tests in ages above XIII and the relatively few tests available in ages XII and XIII? and the further fact that the tests in XII and XIII are admittedly too hard. That the normal subjects, in point of fact, were given fewer tests,2 than the pupils in any other classification appears from table VI. Table VI.?Extent of Advance Testing by the 1908 Scale. Average Number of Full Years Tested Above Basal Age3 Boys Girls Both Average Number of Tests Given Above Basal Age Boys Girls Both Normal… Retarded.. Backward. Deferred.. Borderline. Morons… Imbeciles.. 2.7 3.0 2.5 2.9 3.0 3.1 2.9 2.4 2.6 3.0 3.3 3.2 3.1 2.8 2.5 2.9 2.9 3.0 3.0 3.1 2.8 9.3 11.1 11.0 17.7 13.5 15.2 18.4 6.4 13.5 14.4 19.6 13.1 16.4 18.9 8.7 11.6 12.1 18.4 13.4 15.6 18.6
The average number of ages above the basal age in which all the tests were given was .3 less for the normals than for the imbeciles, who were given the next fewest number (we are using only the averages for the two sexes); and .6 less than for the morons, and .5 less than for the deferred and borderline, who were given the greatest number of complete ages. In the same way, the average number of single tests given to the normals above the basal age was 2.9 less than for the retarded subjects, and 3.4 less than for the 1 Thirteen had a X-year base, 2 a Xl-year, 3 a XH-year and 2 a XUI-year base. * Sometimes due to this lack of higher tests in the scale, and sometimes due to the fact that the diagnosis did not present any difficulty, in consequence of which the examinations were occasionally curtailed because of the limited time at our disposal.
1 Only those ages are included here in which all the tests were given. Because of time limitations it was sometimes necessary to omit the longer tests in certain ages. backward subjects. These three classes were given the fewest advance tests. On the other hand, the groups which were given the greatest number of advance tests, the imbeciles, deferred and morons, were given 9.9, 9.7 and 6.9 more tests, respectively, than the normals were given. From the standpoint of the extent of the advance testing, it would be natural to expect that the scattering should be the greatest among the imbeciles, deferred and morons, and least among the normals, retarded, and backward. We have already seen that this has proved largely to be the case. According to the 1908 scale, the highest average number of advance points were made by the deferred and morons, and the least by the normal, backward, and borderline, while the largest proportion of subjects who earned more than ten advance points was found, again, among the deferred and morons, while the smallest proportion was found among the imbeciles, borderline, and backward. The fact that the deferred cases and the morons pass the highest number of advance tests in the 1908 scale may thus be ascribed to the fact that they were more extensively tested rather than to the alleged fact that they vary more in their mental make up. It is to be noted, however, that the imbeciles, who were given the greatest number of advance tests, scattered over a less wide range than any other group in the 1908 scale?we are referring to those who passed more than ten advance tests?but the imbeciles are followed by the borderline group, which contains cases of potential mental deficiency. As a matter of fact, we have already shown by analysis of experimental data that the groups of subjects who have been extensively tested by the Binet-Simon tests invariably make a higher rating than those who have been less extensively tested.1 This analysis, so far as it went, showed that “the amount of the credit earned depends upon the extent of the testing and not upon the grade of intelligence of the pupils.”
Turning, finally, to the range of ages in which tests were passed in the 1911 scale (table V), the average is the lowest for the imbeciles, morons and borderline, and the highest for the backward, normal, retarded, and deferred. The difference between the extremes, the imbeciles and backward pupils, amounts to only .5 of a Binet age. Possibly the range is slightly exaggerated for the groups which passed tests in the adult age. The range extended to the adult age for 10 per cent of the normals, 14 per cent of the backward (who passed tests in most ages), 13 per cent of the retarded, and 4 per cent of the borderline, while the highest age in which the morons passed any test was XII, the deferred cases Age XI and the 1 Wide Range versus Narrow Range Binet-Simon Testing, Journal of Delinquency, p. 315ff. imbeciles Age X. Nevertheless, the results are in harmony with the findings for the 1911 scale, which Ave have already detailed: The imbeciles and the morons earned the lowest average number of advance credits, and the normal, backward, and deferred the highest average number, while the proportion who earned more than ten advance tests was the smallest for the borderline, morons, and imbeciles, and the largest for the normal, retarded, and backward. Conclusions.
1. The facts revealed by the foregoing analysis do not support the supposition that feebleminded children show more unevenness or greater variation in their mental development than do normal children. On the contrary the variation we have found among our normal children, as indicated by scattering on the Binet-Simon scale, is greater than the variation found among our feebleminded children. We do not consider, however, that this difference is of any considerable significance. We believe that a considerable amount of variation (scattering) in the development of different mental functions is a perfectly normal or typical phenomenon among all classes of human beings. That is, some individuals in any group of individuals, whether so-called normal, abnormal or subnormal, will show a considerable amount of unevenness in the strength of different mental traits, while other individuals will be uniformly or evenly developed. Some normal persons will scatter much, others little. Some feebleminded persons will scatter much, others little. Some insane persons will scatter much, others little. The data presented in this article indicate that the feebleminded do not vary more than the normal, and, therefore, the assumption of Binet and Simon that “unequal” development is the “specific characteristic” of the mentally defective child is not justified. Do the insane and the epileptic vary more than the feebleminded and the normal? None of the original records of our survey of a village of epileptics is now available to the writer, so that an adequate comparison is impossible. But it was the extensive scattering found in our epileptic group of subjects on the Binet scale which first called our attention to this phenomenon. Only a part of our original records for the insane patients are available. These permit of adequate comparison with our normal and subnormal cases, but the conclusions drawn must be tentative, owing to the few insane records available.1 The insane group was given more advance tests than our normal-subnormal group in all the ages permitting of comparison except one (no advance tests could, of course, be 1 Cf. Problems of Subnormality, pp. 157-163. given in either group to those who graded XIII).1 The difference between the averages amounts to 10.6 tests in Age IV, 14.4 in VIII, 8.4 in IX, 1.8 in X, and 1.7 in XI?and 5.5 in II, the sole age in which more advance tests were given to the normal-subnormal group. In spite of this fact, the insane earned only 4.6 tests (average) above the upper base, which is .1 less than for our normal group, and .9 less than for our subnormal and feebleminded groups in Table II. I do not believe, however, that we can accept these figures at their face value, owing to the fact that 28 per cent of the insane passed Age XIII, and therefore could not earn any advance credits, because of the limitation of the scale, while only .47 per cent of the normalsubnormal group passed Age XIII.2 As a matter of fact, the average number of advance credits won by the insane patients who graded below X was 7.3 (median 6.5), which is somewhat higher than for any of the normal-subnormal classifications, while for those who graded X and above, it was only 2.2 (median 0), or 4.8 (median 2) if those grading Age XIII are excluded. Our general impression is that the scattering is slightly greater among the insane and the epileptic than among the normal and the feebleminded. Binet and Simon,3 discussing senile and paralytic dementia, state that the old conception of dementia, according to which it is a “quantitative diminution of the intelligence” or “an injury of all the faculties,” is wrong. They hold that we must replace this interpretation of dementia with the conception of “individual errors of functioning, of defects of every sort, which by their multiplication lower the intellectual level and which present the two following characteristics: irregularity and extensiveness relative to the level of the subjects.” In our judgment more extensive data are needed to establish this thesis. It has not yet been experimentally demonstrated that the mental “irregularity” is so much more pronounced in dements than in other types of the insane or in the epileptic or feebleminded that this phenomenon can be used as a pathognomonic sign of dementia. The amount of irregularity, as measured by BinetSimon scattering, is not greater among the few dements that we have examined than among the feebleminded.
2. In consequence of the above facts it is not safe to attempt to diagnose any one as potentially feebleminded or actually feebleminded on the basis of unevenness of mental development, as evinced by extensive scattering on the Binet scale. It is possible for an expert who is familiar with the entire symptom complex of 1 Cf. Problems of Subnormality, p. 163; and Wide Range versus Narrow Range Binet-Simon Testing,Journal of Delinquency, 1917, Table I, p. 316.
2 Table I, Wide Range versus Narrow Range Binet-Simon Testing, Journal of Delinquency, November, 1917. 8 Binet and Simon. The Intelligence of the Feebleminded, p. 296. special types of mental defectives to make a reliable diagnosis in the case of young children, even though the extent of the mental deficiency at the time is not sufficient to constitute the child feebleminded. But the number of mental deficients who can be thus reliably diagnosed early as feebleminded is probably smaller than has been ordinarily supposed. So far as the majority of cases are concerned we are not justified in making a diagnosis of feeblemindedness, unless we find a pronounced degree of backwardness and unless we are reasonably sure of its permanent character. Feeblemindedness, psychologically considered, means ineradicable arrest of mental development, dating from birth or from early life, of a very serious degree. This is of primary moment; whether or not the child is uneven in his mental development is of secondary importance.
3. The assumption that the essential difference between the feebleminded and the normal is qualitative and not quantitative is not sustained, so far as that assumption is based on the alleged extensive “scattering” or unevenness of mental development among the feebleminded. On the contrary, we believe that the qualitative differences are ultimately resolvable into quantitative differences. When it is said that the normal child has a logical memory and the feebleminded child a rote memory it is not implied that the feebleminded child is utterly devoid of logical memory, so that he is unable to memorize or retain through a comprehension of the logical context. He has logical memory, but far less of it than the normal child. What he has is probably of the same kind as the normal child, but it is weaker in degree.
4. There has been a tendency throughout this and other countries, to which we have called special attention elsewhere,1 to assign so-called potential mental defectives to special classes which were definitely established for “mentally defective” or feebleminded children on the assumption that such children would eventually prove to be feebleminded. In consequence of this tendency, a large proportion of children who are dull, slow or backward, but not feebleminded, have been assigned to classes instituted for, and largely attended by, feebleminded children. We believe that this practice is educationally and socially unwise. The non-feebleminded and doubtful types of mentally backward children should be assigned to ungraded classes, where the work can be more adequately adjusted to their needs. We have elsewhere outlined a plan for the organization of such classes, the practical workability 1 Problems of Subnormality, 1917, Chapter I.
of which we have already demonstrated in some measure, in a large city system.1 5. The assumption that there would be less scattering with the 1911 than with the 1908 scale, because of the alleged greater accuracy of the 1911 (Vineland) scale, is not sustained by our records. The scattering is invariably greater with the use of the 1911 scale in all the classifications in tables I and II, where the figures are based on the averages, the difference in some cases amounting to from two to four advance credits. The cases of extreme scattering (more than ten advance credits) are more numerous in the 1911 than in the 1908 scale, in all the categories, except the borderline, while the cases of least extensive scattering (less than 6 advance points) are decidedly more numerous in the 1908 than in the 1911 scale in all the classifications (table IV). The range of ages in which tests are passed is likewise greater with the 1911 than with the 1908 scale in all the classifications (table V). The advantage of the 1911 (Vineland) over the 1908 scale clearly does not consist in the lessened amount of scattering which obtains in the former scale, for the facts are exactly opposite.
The reason for the increased scattering with the 1911 scale, however, may be the more rigid passing standard used in this scale. All the tests of a given age standard must be passed to get credit for that age in the 1911 scale, while the age standard is passed in the 1908 scale even if one test is missed. The result is that the subject frequently drops to a lower base in the 1911 than in the 1908 scale, while it very rarely happens that the 1908 base is lower than the 1911 base. This is strikingly shown by the figures in table VII, which gives the percentage of subjects who have a lower base (a) by the 1911 scale than by the 1908 scale, (6) by the 1908 scale than by the 1911 scale, and (c) who have the same base in both scales. There are a few instances in each classification in which the 1908 base is lower than the 1911, but the highest ratio of cases in which this is so is only 7 per cent, based on the averages for the two sexes, namely, among the retarded children. On the other hand, the ratio of cases in which the 1911 base is lower than the 1908 varies from 39 per cent, among the deferred cases, to 70 per cent, among the normal cases. In five of the seven classifications the 1911 base is lower than the 1908 for more than half of the subjects. In two classifications the bases are the same for more than half of the subjects, while the percentage having equal bases in the other classifications varies from 27 to 45.
It is evident, from the above figures, that one reason, at least, 1 Problems of Subnormality, 1017, pp. 278-331.
for the greater scattering in the 1911 scale is the more rigid passing standard which causes many of the subjects to drop to a lower base. In other articles we shall compare the scattering in the 1911 Table VII.?Percentage of Subjects whose 1911 Base is Lower than the 1908 Base, whose 1908 Base is Lower than the 1911 Base, and whose 1911 AND 1908 Bases are the Same. Normal… Retarded.. Backward. Deferred.. Borderline. Morons… Imbeciles.. 1911 Base Lower Boys 67 60 61 40 54 40 Girls 80 35 45 37 50 43 59 Both 1908 Base Lower Boys 0 7 1.6 0 1.2 3.2 0 Girls 10 7 2.7 10 0 Both 21 7 1.8 3.6 1 4 4 1911 AND 1908 Bases Equal Boys 32 32 37 60 44 56 30 Girls 10 57 52 53 60 51 31 Both 27 37 40 57 45 54 30
and 1908 scales when the subjects are classified according to BinetSimon age, compute the differences in the mental-age rating by the two scales, and analyze the data for the individual tests. We also expect in future to study the scattering in another scale which we are now using. Thus far we have found just as much scattering in this revision as in the older versions.
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.