A Brief Binet-Simon Scale
The Psychological Clinic Copyright, 1917, by IJfhtna* WHmer, Editor. Vol. XI, No. 7 December 15, 1917 :Author: Edgar A. Doll, Assistant Psychologist, the Training School at Vineland, N. J. I. Many studies have demonstrated the need for employing mental tests in the schools, the courts, the industries, the custodial and corrective institutions, even in the family itself, and in the military establishments of the country. These studies have pointed out that a radical readjustment of social and pedagogical theory and practise must be brought about on the basis of levels of intelligence. School administrators, sociologists, penologists, eugenists, and vocationalists must all consider mental defect and intellectual capacity as vital elements in their dealings with human material.
But how shall society effectively recognize and take advantage of the significance of intelligence, if the means of determining intellectual capacity remain the property of a small group of experts, whose working methods are far too laborious, too complex, and too time-consuming for the practical demands of every-day life? If applied psychology is to make its way in the world and become an instrument in human welfare, its working methods must become the property of the many and must be both refined and condensed. Efficiency is not incompatible with science.
The Binet-Simon Scale has been welcomed as one of the most valuable and economical devices foi* the practical measurement of intelligence. This scale is now available in many arrangements and for all intelligence levels, each particular arrangement possessing its own advantages and its own limitations. But although this scale has reduced mental measurement almost to rule-of-thumb procedures, it must be still further simplified and abbreviated before it can come within the widest ranges of practical utility. Its employment still requires too much time for administration and too high a degree of expert knowledge for the extensive uses to which such a scale might be put. If the Binet-Simon Scale, in any of its forms, 1 The substance of sections I and III of this paper were presented in abstract before Section L of the American Association for the Advancement of Science under title of A brief scale for rapid Binet-Simon examining, at New York, December, 1916.
197 could be abbreviated without serious loss of efficiency as a measure of intelligence levels, there would indeed result a great gain for mental measurement and its many fields of application.
It has long been recognized that the individual tests of the Binet-Simon Scale do not present the same relative degrees of difficulty to feebleminded children that they do to normal children of the same intelligence level. That is, while the individual tests are of approximately equal difficulty for normal subjects of a given age, some tests are characteristically difficult or characteristically easy for mental defectives of that mental age. One may assume, on this hypothesis, that if a scale were devised which employed only those tests which are hardest for defectives, the results from such a scale would tend to exaggerate mental retardation in mentally defective subjects but would leave the mental levels of normal subjects unaffected.1 It would thus be possible to recognize feeblemindedness more easily in rapid mental testing. On this hypothesis I have constructed an abbreviated Binet-Simon scale, made up of individual tests which have been experimentally selected as offering unusual difficulty for mental defectives. This relative difficulty was determined by standardizing the Binet-Simon tests differentially on a group of normal subjects and a group of feebleminded subjects. The experimental data on which this standardization is based consist of wide-range Binet-Simon tests (Goddard revision) of 88 selected normal children from the public school of Yineland, N. J., and 189 feebleminded “children” from the Training School at Vineland, N. J. The test-records of the 88 public school children were selected from records obtained from 250 children by Miss Leila Martin. Two criteria of selection were used, namely, that the records should be complete wide-range tests (that is, should include all years of the scale), and that the children should be “pure” normals, as indicated by an I. Q. range of 90-110. Wide-range tests were necessary in order that the final percentages should be above the criticism of incomplete data; and only strictly normal children were used in order to avoid the excessive variation which is present in groups of unselected children. The number of normal children was relatively small, but this lack is offset by the homogeneity of the group and the completeness and accuracy of the data. This group of subjects was made up chiefly of American and Americanized Jewish and Italian children of ordinary social status. They ranged in life age from 5 to 10 years, and were in the proper school grades 1 This hypothesis is developed and applied by Brigham, from whom I received the suggestion for turning my data to this account. Cf. Carl C. Brigham, Two studies in mental tests, Psychological Monographs, Vol. XXIV, No. 1,1917. I am much indebted to Dr Brigham for many suggestions in preparing this study. Unfortunately I have not had access to the monograph itself, and cannot therefore make specific acknowledgments to his results. for their ages. The feebleminded subjects included all inmates of the Training School whose mental ages were between 5 and 10 years. The average life age for each mental age group was approximately 20 years. The group represents a wide variety of clinical and pathological types of mental defectives. The tests were wide-range tests conducted by members of the research staff.1
Table I shows the classification of data, with the percentages of successes for each test and each age-group. For more direct comparability of results all the normal subjects are classified by mental age groups. Normal age-group 5, for example, includes the normal subjects whose mental ages were from 5.0 to 5.8 inclusive. The Roman numerals in the table indicate the mental year in which each single test is located according to Goddard’s arrangement.2 The successive columns show, from left to right, the age-groups, the number of cases in each group, the average mental age of the group, the average life age, the average intelligence quotient, and the percentage of passes for each successive test of the Goddard arrangement of the B-S Scale. The normal subjects are represented in the upper part of the table and the feebleminded subjects in the lower part. The first line of the table reads: normal children, mental age group 5, 9 cases, average mental age 5.5, average life age 5.4, average I. Q. 101; 78 per cent of whom passed test V-l (compare two weights), 56 per cent passed test V-2 (copy square), and so on. The percentages of passes have been analyzed carefully for influences of examiners’ equation, sex differences, social status, nationality, life age, and previous acquaintance with the tests. This analysis was made after the method described by Brigham (see footnote p. 198). The influence of these factors proved to be greater than the experimental errors which arise from the chance selection of a comparatively small group of subjects.
The relative difficulty of each test of the scale as compared with every other test, and the differential difficulty of each test for normal subjects compared with the same test for the feebleminded subjects was determined by summating the percentages of passes for all successive ages for each test, for each group of subjects. Many other statistical methods were employed for this same purpose, but having proved unsatisfactory their discussion is not pertinent to the present presentation of results. The relative merits of these different standardization criteria will be discussed elsewhere. For present purposes the total sum of percentages indicated in table I is the simplest method, and also yields the most satisfactory results. 1 The statistical tabulation of results for these subjects was prepared for me by Miss Katherine Roese. 2 For descriptive names of the tests consult Goddard’s record blank. All subsequent references are based on this arrangement.
Table I.?Per Cents of Passes on Individual Tests of the Goddard Binet-Simon Scale, for Normal and Feebleminded Subjects. Mental Age Group No. of Cases Av. Mental Age 5.5 6.4 7.3 8.3 9.4 10.4 Av. Life Age 5.4 7.4 8.4 10.4 Av. I. Q. 101 103 99 99 100 Total sum of percentages. Normal Subjects 78 93 100 100 100 100 56 100 100 100 100 100 67 100 100 100 100 100 100 100 100 100 100 100 100 100 100 VI 79 91 100 100 100 67 100 100 100 56 79 100 100 100 100 11 57 78 94 100 100 67 93 100 100 100 100 VII 44 93 91 100 100 100 571 556 567 589 497 559 549 535 440 560 528 444 477 421 486 386 252 341 267 328 33 43 83 94 91 100 22 64 91 100 100 100 0 43 78 100 100 100 22 64 100 100 100 100 VIII 11 7 74 94 100 100 0 14 39 88 100 100 11 7 39 71 100 100 189 Feebleminded Subjects 5.4 6.3 7.4 8.4 9.3 10.4 20.2 18.9 21.0 19.3 20.4 23.1 55 71 90 100 100 100 Total Bum of percentages. 516 Differential difficulty for defectives. -55 58 94 97 100 100 100 549 -7 48 69 97 97 100 100 511 -56 81 100 100 100 100 100 581 52 71 95 100 100 100 518 +21 74 91 100 97 100 100 562 +3 87 94 100 100 100 100 581 +32 74 97 97 100 100 100 568 +33 13 40 77 92 95 100 417 -23 29 60 97 97 100 479 -81 19 57 90 100 100 100 466 -62 45 69 85 97 100 100 496 +52 19 60 69 95 95 100 438 -39 381 -40 32 71 90 87 100 476 31 74 100 95 100 406 +20 217 -35 13 49 90 95 100 100 447 +106 0 0 10 50 71 100 231 -36 13 29 44 55 81 100 322 Mental Age Group No. of Cases Av. Mental Age 5.5 6.4 7.3 8.3 9.4 10.4 Av. Life Age 5.4 6.2 7.4 8.4 9.6 10.4 88 Normal Subjects Av. I. Q. IX 101 103 100 0 0 4 47 91 100 Total sum of percentages. 242 105 237 222 212 0 0 4 18 100 100 119 131 159 74 191 0 0 0 18 73 100 XI 64 79 156 64 XII 39 47 189 Feebleminded Subjects 31 35 39 38 21 25 5.4 6.3 7.4 8.4 9.3 10.4 20.2 18.9 21.0 19.3 20.4 23.1 Total sum o! percentages 169 169 119 I 246 276 139 U lox Aetectwea ^ -V^ y -VtA ?TA \--X7 ?W -yn ?VE -V6 -VV7 --3V \ --20 -V25 ^ -V4 -V38 --n ^ 156 172 126 104 138 95 U 59 25 4 -IS -VSV \ -VTO -V4 85 34
These results make possible an arrangement of the individual tests in a total order of difficulty for either normal or feebleminded subjects; at the same time, they afford empirical measures of the differential difficulty of each test for the contrasted groups of subjects. The relative difficulties of the tests is obvious from inspection. For example, this order is, for the normal subjects, from easy to hard, V-4 (count four pennies), V-l (compare two weights), V-3 (repeat ten-syllable sentence), VI-5 (choose prettier), and so on; for the feebleminded subjects this order would be, V-4, VI-2, VI-3, V-5, and so on. The differential difficulty of each test for the feebleminded subjects as distinguished from the normal subjects is found in the last line of table I; it has been obtained by subtracting algebraically the sums of percentages of each test for normals from the sums for defectives. For example, test V-l is passed by a total sum of 571 for normals and by only 516 for defectives; this test is therefore much more difficult for the defectives. These differential difficulties of the individual tests for defectives are set forth in table II. The numerical values in this table have no intrinsic worth except as comparative measures.
Table II.?Differential Difficulties of B-S Tests for Feebleminded Subjects.
Easy for Defectives Hard for Defectives Patience V-5 + 21 Two weights V-l ?55 Morning or afternoon… VI-1 + 3 Square V-2 ? 7 Use definitions VI-2 +32 “His name is John”…. V-3 ?56 Three errands VI-3 + 33 Four pennies V-4 ? 8 Action in pictures VII-2 + 52 Right and left VI-4 ?23 Verbal comparisons VIII-1 + 20 Prettier faces VI-5 ?81 Days of week VIII-3 +106 Thirteen pennies VII-1 ?62 Better definitions IX-2 + 14 Lacks in pictures VII-3 ?39 Date IX-3 + 9 Diamond VII-4 ?40 Months ! IX-4 +54 Colors VII-5 -10 Money X-l + 37 20-0… …VIII-2 -35 Six digits X-3 + 13 Stamps VIII-4 ?36 Comprehension X-4 + 77 Five digits VIII-5 ? 6 Absurdities XI-1 + 28 Change IX-1 ?73 Complex sentence XI-2 + 6 Five weights IX-5 ?73 Sixty words XI-3 + 17 Design X-2 ?41 Dissected sentences XI-5 + 31 Simple sentence X-5 ?12 Seven digits XII-1 + 20 Rimes XI-4 ?18 Abstract words XII-2 + 25 Long sentence XII-3 + 4 Suggestion XII-4 + 38 Problems XII-5 + 27
Table II gives rise to some questions of grave import for the psychology of mental tests and of mental development. If we assume that these experimental data are at least sufficiently reliable to raise questions, if not to answer them, then it must be obvious that the real psychological contents of individual mental tests are still far from being understood. Psychologists have been altogether too prone to take for granted that the apparent content of a test is the real content, instead of approaching the matter experimentally. Is repeating the days of the week (VIII-3) so very much easier for defectives than normals of the same intelligence levels because of the greater training or experience of the defectives? Then why does counting thirteen pennies (VII-1) prove to be exceptionally difficult for defectives, although it, too, is apparently subject to the same influences; indeed, these subjects have had much more drill on the latter sort of work than on the former. Why should it be that to repeat five digits (VIII-5) proves hard for defectives, when to repeat six digits (X-3) proves relatively easy, and to repeat seven digits (XII-1) proves to be very easy? How shall we account for these qualitative differences between normal and feebleminded subjects? Do there exist fundamental so-called specific faculty defects in the mental constitution of the feebleminded? What is the quality in intellectual brightness which makes practically all of the tests of number comprehension difficult for defectives? Do these data throw some light on the aptitudes of the feebleminded as a class, some indication of those mental differences which, over and above inferior degrees of intelligence, characterize mental defectives? The limited data of this study are neither of sufficient completeness nor reliability to warrant any very serious attempt to answer these questions at this time, but1 certainly they indicate a possible method of experimental investigation and analysis in a very fruitful field. The individual tests of the abbreviated scale have been selected on the basis of the data in table I. This selection is made directly from the experimental data and is not influenced by personal judgments regarding the practical validity of these tests. The selection is based on two considerations, first that each single test should present much greater difficulty of solution to defectives than to normals as indicated by the differential difficulties, and secondly that the selected tests should standardize at some mental year by approximately 75 per cent of passes for the normal subjects, with a steeply and regularly rising standardization curve. Two tests were selected for each year because of the practical convenience of estimating results, and because, as it happened, there proved to be but two tests which could be used at each year according to the criteria of selection. Tests VIII-2 (counting backwards) and VIII-4 (valuing stamps) proved too difficult with time-limits to standardize at year VIII, but when these tests were rescored for all subjects without regard for the time needed to obtain the correct response, then they standardize at year VIII.1
In theory, no standardization percentage is adequate for arranging tests into a scale unless account is taken of the correlations between the individual tests of such a scale. I have not computed these inter-correlations, for I have not been able to find a statistical method suited to the needs of the material. But in this respect this abbreviated scale is no less valid than the complete Binet-Simon Scale, except that in theory the values obtained by a scale made up of two tests per year should be lower than the values obtained by a scale made up of five tests per year. But in fact this theoretical inferiority is denied by subsequent analysis of results obtained by the abbreviated scale. Consequently it appears that the present five tests per year of the complete scale are so highly inter-correlated that more than half of them can be dispensed with, without affecting the reliability of the mental ages obtained.
The selected tests and their arrangement into a year scale is presented in table III. The apparatus, procedures, and scores for administering the tests are identical with those employed by Binet and Simon and as modified by Goddard, with the exception, for reasons already stated, that the time-limits are ignored in scoring the two tests at year VIII. Each test of this brief scale has a mental age value of one-half year. General instructions for giving the tests and for interpreting the gross mental ratings obtained, are the same as are now observed in accepted usage.
Table III?Brief Binet-Simon Scale. V. VIII. 1. Compare two weights 1. Count from 20 to 0 2. Choose prettier faces 2. Count stamps VI. IX. 1- Count 13 pennies 1. Make change 2. Detect lacks in pictures 2. Invent sentence VII. X. 1. Show right and left 1. Give rimes 2. Copy diamond 2. Reproduce design 1 It is interesting to note that although the removal of the time-limits on these two tests affected the percentages of passes for the normal subjects, it did not at all affect the percentages of passes for the feebleminded lubjects.
This brief scale is limited in range from 5 to 10 years, because the experimental data were limited to those age ranges for the normal subjects. But by applying the same methods to groups of subjects of wider ranges of age this scale could be extended in both directions. Six tests, selected on the basis of experience as being probably satisfactory, will be suggested subsequently for years III, IV, XI, and XII. This gives a scale which may be used to advantage in the first five grades of the public schools.
The reliability of the determination of mental ages by this brief scale, as compared with the complete scale, may be measured by correlating mental ages obtained by both scales with the same subjects. For this purpose the complete test-records of all the subjects of this experiment were re-estimated by means of the brief scale of tests. The results yielded a decidedly consistent degree of comparability between the brief scale mental ratings and the complete scale ratings. The Pearson coefficient of correlation between the two sets of mental ratings for the range of ages 5 to 10 was r = .98 (P. E. negligible) for the normal subjects, and r = .90 (P. E. negligible) for the feebleminded subjects. The correlations were also computed for each mental age group instead of for the entire range, but the numbers of cases were too few and the possibility of variation too narrow for the coefficients to be of significance; they ranged around r = .50. From these correlation values, which measure the degree of reliability of brief scale results in terms of complete scale results, one is justified for practical purposes in substituting brief scale ratings for complete scale ratings, by means of conversion constants. These substitutions, as mental ranks, will have practically the same reliability for normal subjects as for feebleminded subjects, but the mental age values for the feebleminded subjects will be from 5 to 10 per cent lower (see table V) by the brief scale than by the complete scale, whereas the mental age values of the normal subjects will be approximately the same by both scales.
This conclusion is sufficiently surprising to merit further analysis. If mental measurement is to extend very far into the fields of pedagogy or of social science, simplicity and economy of technique are secondary only to accuracy of results. Therefore, if two systems of measurement yield results which are so highly correlated that for practical purposes they may be interchanged, it is desirable for practical utility to eliminate the less efficient method, provided that no loss of accuracy ensues. We have seen that this brief scale does retain the reliability of the complete scale as a measure of intelligence level, and in addition has the merits of reduced speed, simplified technique, and reduced apparatus. The technique can be mastered in half an hour by intelligent persons acquainted with the essential principles of educational or psychological measurement, and individual subjects can be examined by this scale in from 5 to 10 minutes. For example, in examining a 9-year-old boy, a mental age of 8.5 was secured in less than ten minutes by means of the brief scale; thirty minutes additional examination together with the tests already administered yielded a Goddard Binet age of 8.2; and twenty minutes of still further examination together with the tests previously administered yielded a Stanford Binet age of 8.4. This case is typical rather than exceptional; it is selected at random. No medical inspector or court official could afford to devote sixty minutes to such an examination under the heavy pressure of hundreds of cases; nor need he do so when he can obtain equally valid results m an examination requiring only ten minutes.
II. The Binet-Simon Scale is made up of five tests per year, and each test is presumed to be located at the year where it is passed by approximately 75 per cent of normal children of that age. The individual tests and the scale as a whole are designed to measure general intelligence. It is assumed that the more angles from which this general intelligence is examined, the better it will be for completeness and accuracy of results. But if one test is statistically as good as another, and if psychologically each test actually does measure general intelligence, that is to say is highly correlated with each of the other tests, then more than one test per year should not be necessary, except, possibly, to avoid chance errors in individual cases.
The statistical reliability of the scale is a function of three variables, namely, the percentages of passes from year to year, the inter-correlations between tests, and the number of tests per year. Up to the present time no one has attempted to determine the reliability of the scale on the basis of these three elements. It has been shown above that a scale composed of two tests per year will give mental age results which correlate almost perfectly with results obtained from the scale with five tests per year. This can only mean that the five tests per year are so highly inter-correlated that more than half of them may be dispensed with without serious loss to the mental age ratings desired.
The question naturally arises, is the brief scale valid because of this conjectural high inter-correlation, or is it due to the innate worths of the tests which have been selected for diagnostic values? Is it essential that the tests should offer exceptional difficulty to mentally defective subjects, or might not a brief scale of tests made up of “non-diagnostic” tests, those which are easiest for defectives, give just as reliable results as one composed of the hardest tests? To test this hypothesis I have composed a second brief scale from the tests which proved easiest for the defectives, and which were passed by approximately 75 per cent of normal subjects for the age where the test is located. This second scale is presented in table IV. In theory this second scale should correlate with the complete scale about as well as the first brief scale, previously developed, for the tests are located according to the same criterion of placement, and, presumably, have about the same degree of inter-correlation. The only difference to be expected is that the mental age ratings obtained by the first brief scale will be somewhat lower for defectives (since all these tests are exceptionally difficult for defectives), and by the second brief scale will be somewhat higher (since all these tests are exceptionally easy for defectives), whereas the mental ages of normal subjects will be approximately the same by both brief scales (since the tests of each scale are for them of indifferent degrees of difficulty). The only differences in the reliabilities of the two brief scales would result from possible unsuspected inequalities in the amounts of correlation between the several tests of each scale. For example, if the difficult tests should be highly inter-correlated and the easy tests not highly correlated, then the first brief scale would prove more reliable than the second.
Table IV.?Second Brief Binet Scale. V. VIII. 1. Distinguish A. m. from p. m. 1. Name days of week 2. Define by use 2. Repeat five digits VI. IX. 1. Execute three directions 1. Give date 2. Solve “patience” 2. Name months VII. X. 1. Describe action in pictures 1. Recognize money 2. Compare verbally 2. Repeat six digits
The actually obtained Pearson coefficients of correlation between mental ages derived by the second brief scale compared with those obtained by the complete scale are r = .95 for the normal subjects (as compared with r = .98 for the first brief scale), and r = .92 for the feebleminded subjects (as compared with r = .90). Apparently, therefore, our conjectures are correct and the easy tests are just about as highly correlated (as a system) as the difficult tests. Consequently, one may expect to obtain almost identical rankings for subjects whether one uses the complete scale or either of the brief scales. The actual mental ages themselves, however, while remaining approximately the same for normal subjects by any of the three scales, would be with defectives lowest by the first brief scale and highest by the second. To convert ratings from one scale to another it is necessary only to apply conversion constants which may be derived from the data of table V. The data of this table are computed for the mental age groups as classified by the complete scale ratings. Mean variations from the averages have been omitted in order to simplify the presentation. These variations are so small (the coefficient of variability was not as great as .10 for any average) that variability is practically negligible. The lowest m. v. was .1 and the highest .6.
Table V.?Comparison of Average Mental Ages by the Three Scales. Binet Age Group Normal Subjects No. Cases Av. M. A. Complete Scale 5.4 6.4 7.3 8.2 9.5 10.4 Av. M. A. Brief Scale No. 1 5.0 6.2 7.0 8.0 9.2 9.7 Av. M. A. Brief Scale No. 2 5.3 6.0 7.1 8.1 9.5 9.6 Feebleminded Subjects No. Cases Av. M. A. Complete Scale 5.3 6.3 7.4 8.4 9.1 10.4 Av. M. A. Brief Scale No. 1 4.7 5.6 6.7 7.7 8.5 9.4 Av. M. A. Brief Scale No. 2 5.8 6.7 7.7 8.6 9.2 9.7
From these considerations it appears that the present BinetSimon Scale may be divided into two scales of approximately the same degree of reliability, and that these brief scales might be used alternately by the aid of conversion constants to correct minor differences in estimated mental ages. One may also conclude that the worth of a test as a reliable measure of mentality is not necessarily to be determined by diagnostic values, but instead is determined by the standardization percentages in relation to the inter-correlation between tests. The advantage of a scale made up of diagnostic tests lies in the automatic exaggeration of mental retardation which such a scale produces with mentally defective subjects. By the use of the first of these brief scales, for example, the mental ages of feebleminded subjects are automatically reduced by approximately 5 to 10 per cent. This may prove to be of very definite service in detecting feeblemindedness in potentially feebleminded subjects and in borderline cases. By the second brief scale the mental ages of mental defectives are slightly increased, which effect has no material value. On the other hand the mental ages of normal subjects are relatively the same by either of the brief scales or the complete scale.
Because of the automatic exaggeration in the mental retardation which the first brief scale shows with feebleminded subjects, the first scale is to be commended to those examiners who must conduct rapid mental examinations intended to yield preliminary indications of mental subnormality. It may also be pointed out that the tests which make up the first brief scale are for the most part the very tests which extended experience has indicated as the best and most reliable from the standpoint of experimental technique. The tests of the second brief scale, on the contrary, are for the most part the least desirable tests of the scale, being too much influenced by chance errors and by mechanical memory.
These brief scales are presented as valid only to a mental age of 9 years (because of the absence of tests in higher levels), and consequently have a limited range of application. They were developed as an experimental demonstration of method; their practical value was not anticipated. But for practical utility I have extended the first brief scale in both directions, by adding tests for the years III, IV, XI, and XII. These additional tests were selected on the basis of my personal opinions as to the most satisfactory tests for those years. These opinions are based on observation and experience with both normal and feebleminded subjects, supported by miscellaneous experimental data. This entire brief scale as now employed for rapid examining at the Vineland Laboratory is presented in table VI. It is to be noted that because of the absence of tests beyond year XII this scale is not reliable in measuring mental capacity beyond 10 years.
A similar record blank for the measurement of mental levels beyond 10 years could easily enough be experimentally developed. I myself lack the experimental data for such an undertaking, but have composed a theoretical abbreviated arrangement of tests from the Stanford Extension and Revision of the Binet-Simon Scale. I have used this abbreviated Stanford scale with very satisfactory returns, but have not a sufficient amount of data yet on hand to be statistically significant. The selection of these tests is wholly conjectural, being based on my personal experiences with the StanA BRIEF BINET-SIMON SCALE. 209 Table VI.?Department op Research, the Training School at Vineland. Record blank for rapid Binel-Simon Testing*
Name Date Mental age Born Time Life age School grade Examiner Status UL VIII. 1. Repeat “His name is John” etc. L Count from 20 to 0 Errors Time 2-GiyeBC*- 2. Count stamps Amount IV. IX. .j _ ^83 1. Make change 20-4 ^ime 1- Repeat three digits 729 25?6 Time 1. 614 2. Compare lines 2. 2. Invent sentence 3. (Philadelphia, money, river) V. X. 6 ? 15 1. Give rime day Time 1. Compare two weights 18 ? 9 mill Time 15 ? 6 spring Time 2. Choose prettier 1. 2. 3. 2. Reproduce design (over) VI. XI. 1. Count 13 pennies 1. Give sixty words 1. 2. Detect lacks Eyes Mouth 2. Arrange sentences 2. Nose Arms 3. VII. XII. 1. Show right and left L. hand Charity R. ear 1. Define abstract words Justice L- eye Goodness 2. Copy diamond (over) 2 gdve problems Hanging from limb Neighbor’s visitors 1 Note.?These tests are to be administered and scored according to the procedures and standards employed by Goddard, except that the time-limit is removed from the two tests at year VIII. The mental age value of each test is one half year, assuming II as a basal year. ford scale with normal and feebleminded subjects. The mechanical arrangement of three tests at some years and four at others is an empirical adjustment of the two-year intervals and the variable values assigned to the upper tests by Terman; this arrangement also retains the numerical ease of estimating results, with each test having a value of one half year in the mental age score. This brief Stanford scale is presented in table VII. It is subject to the same general uses as the complete Stanford scale.
Table VII.?Department of Research, the Training School at Vineland. Record blank for rapid Binet-Simon testing (Stanford revision).1 Name Date Mental age Born Time Life age School grade Examiner Status X. 1. Vocabulary (5 words). 2. Comprehension (2 of 3): Asked opinion Something important. Action vs. words. 3. Designs. XII. 1. Vocabulary (10 words). 2. Dissected sentences (2 of 3): Started for country. Asked teacher. Good dog. 3. Five digits backward (1 of 3): 31879. 69482. 52961. XIV. 1. Vocabulary (15 words). 2. President and king. 3. Arithmetical reasoning (2 of 3): $300. Pencils. Cloth. 4. Problems of fact (2 of 3): Hanging from limb. Visitors. Bicycle. XVI. 1. Vocabulary (22 words). 2. Abstract words (3 of 4): Laziness?idleness. Evolution?revolution. Poverty?misery. Character?reputation. 3. Six digits backward: 471952. 583294. 752638. 4. Inclosed boxes (3 of 4): One large, 2 smaller, 1 inside. One large, 2 smaller, 2 inside. One large, 3 smaller, 3 inside. One large, 4 smaller, 4 inside. XVIII. 1. Vocabulary (28 words). 2. Paper cutting (Binet). 3. Thought of passages (1 of 2): Tests. Opinions. 4. Ingenuity (2 of 3): 3 and 5 to get 7 (begin 5). 5 and 7 to get 8 (begin 5). 4 and 9 to get 7 (begin 4). VOCABULARY. 1. pork. 11. juggler. 21. tolerate. 31. retroactive. 2. outward. 12. regard. 22. artless. 32. ambergris. 3. southern. 13. stave. 23. depredation. 33. achromatic. 4. lecture. 14. brunette. 24. lotus. 34. perfunctory. 5. dungeon. 15. hysterics. 25. frustrate. 35. casuistry. 6. skill. 16. Mars. 26. harpy. 36. piscatorial. 7. ramble. 17. mosaic. 27. flaunt. 37. sudorific. 8. civil. 18. bewail. 28. ochre. 38. parterre. 9. insure. 19. priceless. 29. milksop. 39. shagreen. 10. nerve. 20. disproportionate. 30. incrustation. 40. complot. 1 These tests are to be administered and scored according to the procedures and standards described by Terman in “Tht Measurement of Intelligence.” The mental age value of each test is one-half year, assuming IX as a basal year. It may be advisable to emphasize some of the limitations of the brief scale as well as its advantages. Equivalence in mental age rating must not be misconstrued as meaning complete psychological or clinical equivalence. Neither may one forget that a mental age rating does not in itself alone furnish a sufficient means of mental diagnosis or determinations of feeblemindedness. The more complete measuring scales of intelligence furnish a much greater variety of standard situations in which the subject may be caused to display his mental abilities to the trained observer. Moreover, the results of the more extended examination are more satisfactory by reason of the more elaborate consideration of more phases of the subject’s intelligence, and rule out the possibility of invalidation due to exceptional circumstances of environment or education. The chief consideration is that the gross mental ages and the resulting gross intelligence classifications obtained by use of the brief scales are practically the same as those obtained by the more extended scales. In this capacity a brief scale should prove to be of special value in assisting public school officials to gain insight into the underlying abilities of their pupils. It also furnishes a satisfactory instrument for rapid survey work in intelligence classifications. It is, moreover, a convenient pocket instrument for all field workers and for hasty juvenile court work. Such a scale may be used with accuracy and completeness, because of its very brevity, where a more cumbersome method might be used inefficiently. (To be concluded)
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.