The Individual Tests in the Binet-Simon Scale
- Author:
Wallace Wallin, Ph.D.,
Psycho-Educational Clinic, St. Louis, Mo.
While studying the application of the Binet-Simon scale to epileptic children in 1910 and 1911, we found an “amazing lack of uniformity between the difficulty of the tests of the same age-norms for fully half of the ages of the scale;” we found that scores of our ow and medium grade epileptics passed one or more tests in five, six, or even seven higher age-levels, and scores won from ten to twenty (or even more) points in advance of the first or lowest age in the scale which they were able to pass completely) we found that the B.-S. age given to the child might differ by as much as two or three years according as the rating was figured from the lower or upper base (in 39 per cent of the children the difference amounted to one y ear or more); we found that the “collective difficulty of the different agestandards” varied all the way from .00 per cent of passing to 84 Per cent of passing;1 and we were forced to the conclusion that the v epileptics did “not qualify for a given age on the basis of satisfying the requirements of that age, but on the basis of the system of advance credits from higher ages.” The question thereupon arose whether these anomalous results were due to the unequal difficulty or inherent defects of the individual tests in the different B.-S. ages, or to the wide range method of testing which we employed, or to “fundamental deviations or abnormalities in the epileptic mind. From an anal} sis of the data available then it appeared that the irregularities were not primarily caused by the method of advance scoring from wi e range testing,”2 but were due to defects in the scale itself and peculiarities in the “mental make-up of the epileptic.” Some of the errors which we found in the placement of the tests were confirmed, while others were not confirmed, by the findings of two or three investigators who at that time had tested groups of unselected school children. Since our previous study was made we have had the opportuni y of clinically examining a large number of non-epileptic school children of various grades of intelligence. In order that we might obtain Bmet-Simon records which would be comparable with our records fr?m the epileptic group previously studied we have continued to istered to the epileptics.1 Owing to the exigencies of dispensary work we have not always, however, found it possible to use the wide-range method of testing as thoroughly as we did with the epileptics. The time at our disposal has sometimes been too limited. During the last few years we have also given the new 1911 tests (Vineland version), so as to make it possible to rate the cases according to both the 1908 and 1911 scales. In this series of articles, which will appear from time to time,2 we shall confine ourselves to an analysis of the data for the individual Binet-Simon tests which have been secured from 1000 cases (703 boys and 297 girls), clinically examined in Pittsburgh and St. Louis from the spring of 1912 to January, 1916. These cases are consecutive except that no records are included from subjects who could not be given the B.-S. tests, nor are the records of reexaminations included from subjects examined more than once. Only 16 of the subjects included were over sixteen years of age chronologically, 7 of these being seventeen years old, 1 eighteen, 3 nineteen, 2 twenty-one, 1 twenty-five, 1 twenty-seven, and 1 thirty-five. The intelligence classification of the subjects was as follows:
Boys No. Girls No. %* Both No. Accelerated… Normal Retarded Backward Borderline…. Deferred Morons Imbeciles Idiots Feebleminded. 52 87 280 85 17 94 73 7 174 1.1 7.3 12.3 39.8 12.0 2.4 13.3 10.3 .9 24.7 14 21 22 94 26 16 58 44 2 104 4.7 7.0 7.4 31.6 8.7 5.4 19.5 14.8 .7 21.5 22 73 109 374 111 33 152 117 9 278 2.2 7.3 10.9 37.4 11.1 3.3 15.2 11.7 .9 27.8
1 See ” Experimental Studies of Mental Defectives,” p. 119f. In giving test No. 10 we have never asked “What is it about?” This question was inadvertently introduced into the Guide. We did not have all the sentences in tests 8, 21 and 58 standardized at the time some of the epileptics were tested. For test 23 nine-year credit has always been given for mere descriptions, while the rating has not been based on the replies to the supplementary question given in the Guide, when a classificatory definition was not spontaneously proffered. In test 43, we have used 25-6c. In 46, the weight test, we have always used metal pill boxes of 6, 9, 12, 15 and 18 grams, similar to those used at Vineland. We have, however, never suggested to the child to revise his arrangement, as suggested in the text, except when it has been apparent that an impulsive child has gone through the operation in a wholly perfunctory manner. These instances of suggested revision have been very few. We have never arranged “the weights while the subject looks on,” as stated by Terman. After the subject has arranged the boxes we have merely turned them upside down, retaining the same arrangement. We have asked the subjects to use one hand when two have been used.
2 The general practice of requiring authors to subsidize the publication of their experimental monographs in psychology, the revenues from the sales of which rarely enable the author to recover his initial investment, makes it necessary to utilize the medium of the periodical press. At the present time we are more in need of grants for the publication of original investigations than for the prosecution of researches. 8 Based on the total number of boys (703). 4 Based on the total number of girls (297). ‘Based on the total number of boys and girls (1000).
Only 27.8 per cent of these subjects were diagnosed as feebleminded. Those who roughly classify children into two categories, normal and feebleminded, would have classified a considerable majority of them as normal. In analyzing our mass results it should be remembered, however, that some of our subjects were feebleminded. Only twelve of the subjects were epileptic (including one case of masked epilepsy ), which is so small a proportion of the entire number as not to affect the general averages.
In this article we shall confine ourselves to a consideration of the data for the B.-S. weight discrimination test placed in Age IX in the 1908 scale and in Age X in the 1911 scale.
The Weight Discrimination Test.
Our interest was attracted to this test in 1910 because of the very unsatisfactory results which we obtained from its use on epileptics, when we strictly adhered to Binet and Simon’s method of administering the test. When administered in the classical way, success in the test depends not only on the ability to discriminate small differences in weight by the kinesthetic sense, but also on the ability to follow verbal directions. But when so administered it is frequently impossible to determine whether success in the test depends primarily on keenness of muscular sensitivity or on ability to understand the directions. In preliminary trials on epileptics in 1910 we found that the test was entirely too difficult for epileptics who graded in the B.-S. age in which it had been placed, or in any of the higher mental ages Some of these obviously failed because they were unable to distinguish the difference between the weights, but many failed because they were unable to comprehend the instructions. We found, however, that many of the latter succeeded when once the directions were understood. Moreover, the only way of getting many epileptics to understand the test was to demonstrate the method of procedure. Many were unable to execute the test from repeated instructions according to the B.-S. procedure, so that almost in despair we adopted the procedure described in our Guide (pp* 123 and 133). We usually only hefted three or four of the boxes, but as stated on page 80 we have never placed them in position.
Under this method of administration we still found the test entirely too difficult for the IX-year epileptics, only 44 per cent of whom succeeded in the test, while it has also proved too difficult for the IX-year olds in our non-epileptic group, only 59.4 per cent of whom passed the test (based on 133 IX-year records). The results 1 We have not followed Lombroso’s conception of epilepsy.
are somewhat discrepant among other investigators who have tested “normal” or unselected children, only some of whom have adhered to the original method of administering the test. Decroly and Degand concluded that it belonged in the fifth or sixth year, while Goddard retained it in age IX, because 80 per cent of 55 nine-year olds succeeded, while singularly only 40 per cent of 40 eight-year olds succeeded. The record for his ten-year olds is based on only four cases and cannot, therefore, be used for comparison. On the other hand, only 30 per cent of Johnstone’s, 55 per cent of Terman and Child’s, and 60 per cent of Bobertag’s nine-year olds, and 78 per cent of Bobertag’s ten-year olds passed the test?the number of cases examined by each of these investigators in these ages being only 30, 49, 40, 32, respectively. Binet and Simon placed it in age X in the 1911 revision. In several of the above investigations, therefore, the test has proved too difficult for Age IX, on either the 60 per cent or the 75 per cent standard of passing.
Terman, who retains the test in Age IX in his latest revision, prescribes a different procedure for the two boxes in Age V than for the five boxes in Age IX.1 He does not approve of our method of illustrating the procedure, on the ground that the “experiment was not devised primarily as a test of sensory discrimination, for it has long been recognized that individuals who have developed as far as the 9- or 10-year level of intelligence are ordinarily but little below normal in sensory capacity.” He regards it rather as a test of ability to comprehend instructions, to keep an objective in mind and find the means of reaching the objective. He apparently feels that the test does not correlate very closely with “true mental age.” We have analyzed our data on this point and find, on the contrary, a very significant correlation with mental age (B.-S.),as shown by the percentage of successes in each B.-S. age from VI to XIII:
B.-S. Age Boys. Girls. Both. VI No. % 10.5 20.0 12.5 VII No. % 28.7 11.7 23.3 VIII No. % 40.4 28.3 36.5 IX No. % 65.2 44.7 59.4 No. % 68.0 57.9 65.0 XI No. % 76.1 71.4 75.3 XII No. % 100.0 66.6 83.3 XIII No. % 70.0 100.0 72.7
1 Our percentage of successes for the two boxes in Age V was 66 per per cent for the V-year old epileptics (based, however, on very few cases), and 92.8 per cent for the V-year olds in the non-epileptic group, being 96 per cent for the boys and 87.5 per cent for the girls (based on 42 boys and girls). According to our results for the non-epileptic group the test is too easy for Age V, a finding in harmony with the conclusion of Petersen and Doll: “Sensory Discrimination in Normal and Feebleminded Children,” Training School Bulletin, 1914, November and December This article gives the latest summary of experiments on weight tests.
There is a decided increase in the percentage of successes in every ascending age except XIII?the data in the two upper ages, however, being based on only a few cases. The increase is particularly marked in the lower ages. The amount of increase in the percentage of passes between the successive ages from VI to XII is: 10.8 per cent, 13.2 per cent, 22.9’ per cent, 5.6 per cent, 10.3 per cent, and 8 per cent. The increase holds fairly uniformly for both the boys and the girls. ?sot only so, when the data are averaged separately for each intelligence category, we likewise find a decided increase in the ability to perform the test with each higher category of intelligence, with only one exception, as shown by the following percentage of successes:
Boys. Girla. Both. Imbeciles No. % 14.2 20.0 15.7 Morons No. 83 52 135 % 36.1 28.8 33.3 Borderline No. 57.1 34.6 51.0 Backward No. 236 79 315 56.2 44.3 53.3 Retarded No. % 52.8 40.0 50.5 Normal No. 39 14 53 % 61.5 35.7 54.7 Accelerated No. % 75.0 80.0 78.5
This is a strikingly significant increase in view of the fact that many chronological ages are contained in each intelligence category. The increase between the morons and the borderline amounts to 17.7 per cent.- It is evident from our results that the test, as we have given it, is not purely a test of sensory discrimination?nor can it be administered purely as a test of sensory discrimination?but also of intelligence. Success in the test, even when given primarily as a sensory test, depends upon two factors: first, the subject’s keenness of sensitivity and, secondly, his grade of intelligence. This is merely a specific instance of the assumption, justified on other grounds, that there is a specific and a general factor involved in any mental experiences, the specific factor being correlated with the functioning of a specific cortical area and the general factor corresponding to the functioning of the adjacent or associated areas and the general tension or activity of the cortex as a whole. The ability to perceive differences in the weights postulates the ability to attend, to retain dental impressions of the weights that have been hefted, to compare these impressions, and to keep the task constantly in mind as the weights are being hefted?certainly no insignificant intellectual problem. In fact, success in the test, when administered as a sensory test, depends more on central than peripheral factors. We find, therefore, in harmony with this interpretation, that the normal or backward pupils do better than the morons, and the morons better than the imbeciles in this sensory test. In a series of other sensory and perception tests on epileptics and normal children the results point to the same conclusion (described in a monograph on “The Measurement of Mental Traits in Normal and Epileptic School Children”). Similar results have been obtained by Wiley and Norsworthy, while the results of both of these investigators, however, agree with our own in showing that the difference is greater for complex than for simple traits. Binet and Simon’s conclusion that the imbeciles’ fineness of perception equals or nearly equals that of normal persons was based only on three defectives and only one normal, evidently belonging to the lower grade of normals. Petersen and Doll in a different experiment on the discrimination of weights found that the feebleminded were slightly more obtuse than the normal of the same mental age, but this is ascribed to intellectual rather than to sensory factors. We do not think it is possible in the way this weight test is administered to isolate, except theoretically, the sensory elements in the complex, so as separately to measure the elemental sensory experience. In practice the intellectual and sensory components are integrated or fused and can be measured only as a unitary whole.
Incidentally, it may be pointed out that we have found both imbeciles and morons who have been able to pass the test. Bobertag had said that no child of eight or nine who performs the test properly can be feebleminded?and his method of giving the test approximates our own. We are less sanguine. We do not believe that the passing of any single test will prove any child to be normal, nor that the failure on a single test will prove any child to be feebleminded, except in perfectly obvious cases.
Conclusions.
1. Success in this weight test depends both on kinesthetic and on intellectual factors, even in the simplest practical form in which it can be administered.
2. The successful performance of the test increases with ascending B.-S. age and ascending intelligence classification. It is possible that if the kinesthetic elements could be measured wholly independent of the intellectual?i. e. as elemental sensory experiences?there would be no such increase or that it would be greatly reduced. But the increase occurs even when the test is administered in its simplest practical form. This being the case,
3. It seems unwise to complicate the test by making the directions difficult to follow?by converting the test into a difficult directions test: the test is already difficult to perform because of the small differences between the weights, and the ability to perform the test already varies with mental age or intelligence.
4. Specifically the test is sufficiently difficult for Age IX, without the addition of the difficult “directions test” feature. Indeed, we are not convinced that it is not too difficult for this age there is as yet no agreement on the percentage of passes which should constitute a test normal for a given age.
5. The improvement in the test with increasing mental age or intelligence?also shown by Petersen and Doll in their test on feebleminded and normal children?would seem to indicate that its use as an intelligence test is legitimate. It happens, of course, that some bright children of considerable maturity occasionally fail on the test, while imbeciles may succeed.
6. In our results the boys surpassed the girls in all the B.-S. ages and in all the intelligence categories except two. In our previous test of epileptics the boys, again, surpassed the girls. These boys, however, averaged somewhat higher in intelligence than the girls. The findings among other experimenters are somewhat discrepant.
7. Our earlier1 conclusion that epileptics “apparently suffer from a blunting of the kinesthetic sensitivity” seems to be reinforced by the data presented in this article, for the percentage of successes in the test was less for the epileptics than for our non-epileptic group of the same B.-S. age. Naturally our results for the epileptics are less reliable because the number examined was less. 1 As before, p. 54.
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.