The Binet-Simon Tests in Relation to the Factors of Experience and Maturity
- Author:
Wallace Wallin, Ph.D.,
Psycho-educational Clinic, Board of Education, St. Louis, Mo. In the December (1914) number of The Psychological Clinic, Messrs. Pintner and Patterson supply data, from an analysis of the Binet-Simon records on file in an institution for the feebleminded, which indicate that the adult inmates (“above fifteen”) are superior to the child inmates (“below fifteen”) in the two tests studied, namely, the naming of the days of the week and months of the year. The results show that on the basis of a passing standard of 75 per cent, the days-of-the-week test is normal for the adult group at the B.-S. (Binet-Simon) age VI and for the child group one year later.
On the other hand, the months-of-the-year test is normal for both groups at age VIII. That the adults’ superiority compared with the children’s is actually larger in the months tests than in the weeks test, instead of being smaller as would appear from the above statements, becomes apparent if we compare the average efficiency for all the B.-S. ages of the two groups in the two tests. The average per cent of passes (the general averages of all the age averages) in the months tests is for the adults 56.2 and for the children 36.3, a difference of 19.9, while the corresponding averages in the days test are 76.3 and 63.6, a difference of only 12.7. That the records, when thus compared, show that the adults did relatively better in the months test would, I believe, be in accordance with expectation, on the authors’ theory that the adults’ superiority in the two tests is due, not primarily to the fact that they have attained a higher stage of intelligence, but to the fact that they have had more experience with the tests in question?the experience that comes inevitably with the necessity in workaday life of attending to the days of the week and months of the year.
Several years ago a similar analysis of “maturity” difference1 was made for various tests, based on the testing and the B.-S. classification of subnormal epileptics. A comparison was made between the records of children and adults for the following tests: time of naming the four B.-S. colors, time of reading the B.-S. selection, number of memories retained from reading said selection, time of uttering 1 Called maturity difference simply to indicate “differences between the juvenile and adult periods of life.” Experimental Studies of Mental Defectives: a Critique of the Binet-Simon Tests and a Contribution to the Psychology of Epilepsy, 1912, p. 69. 60 detached words, time of replacing blocks in the form board (Vineland modification), and the strength of grip. The “adults” included those twenty-one years or over, and the “children” those who were less than twenty-one. Twenty-one was chosen as the dividing line, not especially because it is the point of differentiation in common law between the period of minority and legal majority, but because the processes of physical maturation are frequently considered to be largely completed at that age, and especially, according to my observation, the mental differences due to repetition, habituation, experience, become apparent appreciably later in subnormal than in normal individuals. Hence it was believed that significant differences in the B.-S. rating, or in the results of individual tests, between the groups of epileptic adults and children as thus defined would probably represent differences dependent upon maturity and experience. While no attempt was made to justify this construction from an elaborate analysis of the results of the individual tests which were studied, a mere inspection of the curves of mental age distribution showed that the adults graded decidedly higher than the children. There were less adult than child idiots and imbeciles, and more adult morons. The conclusion seemed to follow that “as the B.-S. scale is now constituted, it may be assumed that defective adults will grade somewhat higher than defective children.” … “The adults among defectives possess a larger storehouse of experience and acquired knowledge.2
On the other hand, the analysis of the tabular data for the six individual tests3 which were specifically investigated revealed significant maturity differences in some tests but not in others. I shall here give the results of a more complete analysis of the first five tests mentioned above than was possible in the earlier treatment. The adults clearly excel in the reading test. They read the selection in less than half the time (47.6 sec.) required by the children (97.4 sec.); and they vary about half as much among themselves in the time required to read it (M.V., 14.3) as do the children (M.V., 26.6). The children, however, gain almost three times as much from B.-S. age to B.-S. age between IX and XI4 (average improvement per year, 37.2 sec.) as do the adults (13.8 sec.).
A surface examination of the results for the form-board test (table XIII in the book) would indicate that the adults surpassed the children considerably in the speed of the performance (47.4 sec. 2 Ibid., p. 15 and Table I. 3 Ibid., pp. 69, 72, 73, 79, 82, 88, 97, 99. 4 The comparisons must be limited to certain ages throughout because of lack of comparative data for some ages. 268 THE PSYCHOLOGICAL CLINIC. as against 65.2 sec.) and in the amount of improvement with each mental age from V to XII (14.4 sec. per B.-S. year as against 7.7 sec. for the children), although their M.V. was larger (17.1 vs. 12.9). An examination of the table, however, indicates that the average for the children is vitiated by the dominant influence of an unusually large score in age III. If this is eliminated the difference between the general averages will be reduced to 3.9 sec., and the children will excel (31.3 sec. vs. 35.2). In all the other tests the children excel in the average scores, in the amount of improvement with increasing B.-S. age and sometimes in the M.V., but the differences in the average efficiencies are not large. Thus it takes the children on the average .8 sec. less time to name the colors than the adults, while their M.V. is only .2 less than the adults’. They utter only .9 of a word more in 3 minutes than do the adults (38.9 vs. 38.0 words), while their M.V. is .1 larger. Their average amount of improvement from mental age to age is, however, more considerable (11.8 words vs. 8.6 for the adults). The children reproduce from the reading selection .6 units more than do the adults (5.5 vs. 4.9 units), their M.V. is .3 less (1.1 vs. 1.4), while their improvement with each successive B.-S. age is .6 unit more than for the adults (1.8 vs. 1.2 units).
It is thus evident that there is a significant difference in only one test, or at least two: in the time to read the selection, in which the adults excel, and in the form-board test, in which the children excel. The greater facility of the adults in reading is explicable on the assumption that they have had more practice in reading than the children, although the institutional type of epileptic adults are not notorious for the exercise of their reading propensity. The form-board test, on the other hand, represents a novel situation. There are practically no situations in the daily life of the institutional adult epileptic (or of the average normal adult either) which call for the form-board type of response. Hence ordinary experience fails to provide any specific practice for this performance?specific practice for the specific type of skill required. It would only affect the general factor; that is, intelligent adaptation to a novel situation. It is possible, however, that children, both in and out of school, receive a certain amount of specific training for this performance through their construction games with blocks of various shapes, and through their plays with puzzle boards somewhat similar in nature. Most of the epileptic children tested were school cases. This may explain the slight superiority of the children in this test. Similarly, the fact that the children named the colors slightly more rapidly and reproduced slightly more memories from the reading selection than the adults did may be due to the fact that they had been given practice in school in the naming of colors and in the reproduction of contents of selections read. However, our results show a significant difference in only one test, and therefore suggest that it is legitimate to use the same mental age norms for both children and adults in all the tests except the reading test.5 This conclusion, of course, needs to be verified for other types of subnormals than epileptics (and forsooth, also for normal children and normal adults) when arranged into similar groups: children under twenty-one years, adults twenty-one years and over.
If the dividing line between the children and the adults is drawn before the children reach maturity, the adult superiority in a given test might depend upon the factor of maturity rather than on the factor of experience. The fact that the epileptic children improved relatively more with each ascending B.-S. age than the adults in all the tests would seem to indicate that not only the mental age but also the chronological age of the children increased with each ascending B.-S. age. The data are not now available for demonstrating the correctness of this supposition with the present group of children, but it has been found to be true for a parallel set of subnormal juvenile cases. It should follow therefore that the children gain more than the adults from mental age to mental age not because their mental age increases, since there is a parallel increase for the adults, but because their chronological age or maturity increases. With the adults on the other hand, there would be no growth in maturity, even if the chronological age increased slightly with increasing B.-S. age, on the supposition that the adults had practically attained their maturity at the age of twenty-one.
The fact that the children gain more with ascending B.-S. age would suggest that there should be less difference between the children and the adults in the higher B.-S. ages, than between the children and the adults in the lower B.-S. ages. In order to test this supposition I have worked out the amount of difference between the children and the adults, first for the children’s and adults’ averages for the lower half of the B.-S. ages, and, secondly, for the children’s and adults’ averages for the upper half of the B.-S. ages which are tabulated for each test.
Actually the difference between the children and adults was larger for the lower B.-S. ages in four tests: the four colors (the difference was larger in the lower ages than in the upper by an average difference of .84 sec. per B.-S. age); the form board (larger by an ? But facts have been given tending to show the validity of the reading test as a test of intelligence: as before, p. 72. Granting that this is so, separate norms for children and adults may still be required in this test.
average difference per age of 23.9 sec.); the reading selection (by an average difference per age of 45.6 sec.); and the week days (by an average difference per age of 13.0 per cent; computed from Pintner and Patterson). In the first two tests the children excel, while in the last two the adults excel. On the contrary, in the three remaining tests the differences between the children and the adults are larger in the upper B.-S. ages: 60-words (by 5.5 words per age); memory units (by 1.90 units per age); and months test (by 9.2 per cent per age; computed from Pintner and Patterson). In the first two tests the children excel while in the last the adults are superior.6 The results are thus only partially in harmony with the supposition. What is the explanation? The suggestion is, first, that the difference between the children and the adults is larger in some tests in the lower B.-S. ages because in those tests there was a larger per cent of children tested relatively to the per cent of adults in the lower ages. Secondly, there should be conversely a smaller percentage of children relatively to the number of adults in the upper B.-S. ages for those tests in which the difference is larger in the upper ages. The first supposition holds so far as the form-board and weekdays tests are concerned. The percentage of children is 10 per cent larger than the percentage of adults in the lower ages for the form-board test; while for the week-days test it is 11 per cent larger. It is noteworthy, however, that while the children excel in the form-board, the adults excel in the week-days test. In the reading test the percentages are the same, while in the color-naming test, in which the children excel, the per cent of adults in the lower ages exceeds the children by 5 per cent. The second supposition is confirmed for all tests. However, there are only 2 per cent more children than adults in each test, and while the children excel in two of these tests (GO words and memory units), the adults excel in the third (months). Our analysis thus leaves the discrepancy unexplained, and points to the necessity of a further study of the effect of the factors of maturity and experience on large numbers of different types of subnormal persons (as well as normal individuals), in order that we may know more surely whether it is necessary or desirable or possible to exclude from intelligence scales those tests which are influenced by experience, merely because they are so influenced and irrespective of whether or not the tests in themselves are intrinsically valuable or give us an insight into traits which are significant for mental or social growth and development.
From the earlier analysis of the B.-S. results it was conjectured ? It should be said that in all tests when the children are superior to the adults in the lower B.-S. ages they are also superior in the upper ages. The same statement applies when the adults are superior to the children.
that it is not “essential to eliminate (from scales of development) all the tests which are dependent upon training … partly because this is not desirable and partly because this is impossible. Nature and nurture proceed hand in hand, inseparable, reciprocal, interacting and independent only in conception. Just as we posit a normal rate of development which the forces of human nature undergo?normal, that is, within limits?so we may posit a normal curve of development within variable limits for a given order of civilization or social evolution, which human changes follow as a result of the processes of nurture. We cannot, if we would, test merely pure native capacity uninfluenced by environmental agencies, except possibly during the first months of life. But we can measure native capacity as modified by the environment” (p. 57).
The analysis of a number of individual tests in this paper seems, therefore, to show that common age standards can be used, without very much error, for both subnormal children and subnormal adults, in all the tests except in reading and naming the week days (and possibly naming the months of the year), where the mental age scores are lower for the children than for the adults. However, whether the line between the juvenile and adult periods of life should be drawn at fifteen, twenty-one, or some other age, will depend on whether maturity is attained at fifteen, twenty-one, or some other age; and this can be ascertained only by experimentally determining for separate mental traits the age beyond which they fail to show any increase in strength or functional capacity.
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.