Performance Tests with Defectives
- Author:
D Lucy G. Fildes, B.A.,
Research Student, Psychological Laboratory, Cambridge.
There is no need, in writing for this journal, to emphasise the service which the *Binet-Simon Scale for the Measurement of Intelligence and its many revisions have rendered in the study of mental deficiency. The substitution of objective measurement in diagnosis for mere personal opinion ; the possibility of estimating degrees of defect and of stating them in standardised language; the realisation that a condition of lowered intelligence does not as a rule improve with age; all these but rarely outline its benefits. And yet, perhaps because of the very vital importance of its services, we have at the present time great need to pause and consider lest in our use of the Binet Scale we attribute to it powers which it does not and cannot possess. For in dealing with the individual the apparent ease with which it estimates intelligence may prove our undoing. Mental ages and intelligence quotients may become magical formulae giving a spurious appear- ance of knowledge and leading to inferences entirely unjustified in fact and disastrous in application.
In reality, as a moment’s thought will reveal, a mere acquaintance with the mental age and intelligence quotient of a child tells us singularly little about him. While serving its purpose in diagnosis, in classifying him broadly, that is, with a certain group of his fellows for practical ends, it alone forms a quite inadequate foundation for educational treatment. For to educate truly we must have an intimate knowledge of each individual’s peculiarities, and cannot be content with the bare bones a mere statement of mental age supplies. Especially is this so in working among the sub-normal, whose demands on the educator are far greater than are those of their normal fellows. Indeed, even though the concepts of mental age and of intelligence quotient be made more explicit and therefore of greater value by an exact statement of the results of the different tests attempted in each case, the limitations inherent in the nature of the Binet scale itself will prevent their ever being, if taken alone, adequate for sound educational and social prognosis. They are valuable, but something more is needed.
As a matter of fact, we have to admit that in certain types of cases their very value is in doubt. A Binet mental age and intelligence quotient may, far from being helpful, be positively misleading. For a fair application of the scale demands that the individuals tested shall have had at least average educational opportunities. Consequently, an abnormal external environment, on the one hand, or a special and limited mental disability, on the other, may result in a false suggestion of general deficiency. Cases such as those of canal boat children, cut off so largely from the ordinary experiences of modern life; and of word- blind children, who suffer mainly from failure in visual discrimination, are here in point.
But apart from these special cases we are coming to realise more and more the need for an exact knowledge of the defective greater than that which can be given by the Binet scale. Obviously, the more such knowledge approaches in type that given by the scale, in other words the less it depends on the mere opinion of the individual investigator and the more rapidly it can be gained, the greater is its value likely to be. Most urgently is it needed with the higher grade defectives, in whose case defect or social competence is determined far less on a * Throughout this Article the term “Binet Scale” should be read as including the many revisions of the original scale.
basis of simple intelligence than on the possession of certain temperamental and emotional characteristics ; for it is these which really decide the use that can be made of the intelligence possessed. Any attempt to diagnose defect on a basis of mental age or intelligence quotient alone is psychologically indefensible. It assumes a division of mental power which does not take place since the operations pf the intelligence are necessarily bound up with other mental activities. True, individuals of slightly subnormal intelligence seem very frequently to suffer also from peculiar emotional instabilities, which render them socially and industrially incapable under the conditions of modern civilised life; and so far no entirely adequate reason for this fact can be supplied. But there are also many, definitely subnormal in intelligence, who, with reasonable help and education, can make good in society in spite of their disabilities : provided only, that is to say, that their intelligence is above a certain minimum.
In distinguishing between these two groups, i.e., between those high grade defectives who are likely to be socially competent and those not so competent, a mental age or intelligence quotient on the Binet scale is of little, if any, help. A boy may, for instance, attain an intelligence quotient of 70 only, or even less, and yet being slow, plodding, good-tempered and governed largely by routine and able to stick to a job, be capable of supporting himself with com- parative ease in a suitable occupation; while another, better in intelligence but careless, flighty, irresponsible and readily distracted from work will never make good in the world and will always need special care. It is because of this inadequacy in the Binet scale that at present the decision as to the adult certifica- tion of these high grade children is influenced so much by the opinion of those responsible for their education, as indeed up to a point it should and must always be; seeing that nothing can replace prolonged study by experienced observers. But the value of standardised tests, if such are available, as supplementary dis- criminative evidence is obvious enough; especially when ffie Binet mental age is very high. Personal opinion would then receive a backing demonstrably free from prejudice.
Such tests would demand that the child’s intelligence be carried over into some form of practical activity not yet learnt; that he be given a chance to show whether he can and will carry on under difficulties and whether he can adapt himself to new situations without undue loss of time. They would further permit of the comparison of his results with those of others. These requirements are fulfilled by the now rapidly increasing group of tests known as performance tests. Performance tests satisfy in the first instance the essential requirements of all mental tests. They are standardised in procedure and in marking. They Permit comparison of individual results with the average results of large groups. In other words, they constitute, when rightly used, objective measurements resembling those of the Binet scale. They have, however, certain definite Peculiarities. Although demanding as a result an actual doing of something, they are planned to measure intelligence and not manual dexterity alone. The mere act is never what matters, but rather the ability to discover what to do and to do it : often to do it in spite of many failures. Planned as a rule to supple- ment difficulties inherent in the Binet scale, performance tests demand a minimum of language. The words of the investigator in giving the necessary diiections are few and simple. The subject need not speak at all. Although affected, as are all our actions, by past life, they are less influenced in their results than are most of the tests of the Binet scale by the normality of past experience. Absence from school, for example, will affect them comparatively little, while, on the other hand, ordinary school teaching will not influence them much. One must bear in mind, however, a fact often forgotten : that, from the results of one 90 Mental welfare.
performance test only, no conclusions whatever can be drawn as to the capabili- ties of an individual. The nature of any one test, or the circumstances of its administration, may always be specially favourable or unfavourable. A result may be gained by chance?this is very hard to guard against in some tests? or it may actually have been known beforehand. And, anyhow, the data are inadequate for inference.
Consequently, for anything approaching a fair estimate of capacity many tests must be used. At once arises the difficulty of the time taken for testing. But, after all, time is necessary for good work. We cannot misjudge the children simply because we are in a hurry; and it is only in difficult cases that performance tests will be needed for actual testing purposes.
What has been said already suggests that performance tests will be most useful with certain types of individuals, with those, that is, having abnormal past experience or suffering from specific defects. It seems fairly obvious that a boy eleven years of age, not deaf, yet unable to use or to comprehend language, who attains a mental age of twelve and a half years when tried on ten performance tests, is not mentally deficient in the ordinary sense of that term; nor can he satisfactorily be educated with mentally deficient children. Yet, apart from performance tests, it would be hard to find any satisfactory evidence of his intelligence. We may know that he is intelligent, but without definite facts showing how exactly he compares in a variety of ways with his normal fellows we cannot prove this. So with the child who cannot learn to read. Technically, he is defective. He cannot ” profit by the instruction given in the elementary school.” But performance tests will help enormously in determining how best to teach him. For a mere inability to read will not make him defective in adult life, and his education must look forward to the future. Other specific disabilities, influencing school work in particular and reflected in Binet results, especially with older children and adults, may reveal themselves in the same way. It is very possible that many children, judged subnormal in school and on Binet results, are not subnormal at all in what one may call practical ability, and may be saved from the despairs and disasters of constant failure if only this is realised in time, and education fitted to individual needs.
In the same way, performance tests give a chance of trying out children of unusual early life. The Binet tests are confessedly based on the assumption that the individuals doing them have had certain common bases of experience and opportunity and that native intelligence displays itself in ability or in failure to learn from these experiences. But children brought up mainly in isolation with uneducated parents, such as canal boat children, have not had opportunities other- wise almost universal in civilised life. For them, therefore, a Binet examination is not a fair test. Performance tests may not be altogether fair either, but they do make lower demands on the use of knowledge which must be derived from the external environment.
Let us then assert definitely that performance tests are of use in some cases of difficulty when the diagnosis of defect is in question. If only, that is, enough time can be spared for the work. But their value in the educational and social prognosis of high grade cases is less certain and needs a closer investigation. *Porteus, in discussing his Maze Tests?the most used probably of all perform- ance tests?makes for them very definite claims in this direction ; maintaining that they reveal certain important temperamental characteristics, especially the habit of prudent preconsideration. He shows further that their results correlate highly with his ratings of industrial capacity and of social adaptability. ” Com- bined with a Binet examination,” he says, ” they make diagnosis much more * Guide to the Porteus Maze Tests, by S. D. Porteus, 1924.
certain.” True, he discounts similar claims made by other performance tests and scales; but even he would doubtless admit that a scale of performance tests including” the Porteus mazes was of some diagnostic value as a supplement to, rather than as a substitute for, the Binet scale; particularly when the social and industrial adaptability of high grade defectives is in question.
Of interest here are the results of the examination of forty-eight high grade boys on such a scale. The scale used was that described in ilhe British Journal of Psychology, April, 1925. It consists of fourteen tests, chiefly American in origin, but restandardised on English children, the Porteus mazes being included. 1 he boys examined varied in Binet intelligence quotient from 34 to 93, only five of them having I.Qs. below 50, and fifteen having I.Qs. above 70, i.e., being definite problem cases. On the performance scale the I.Qs. varied from 36 to 108, twelve being below 50 and twelve above 70. The performance I.Qs., that is, as worked out on this scale, tend to be lower than the Binet. But, seeing that we are dealing with two entirely dissimilar sets of tests whose results are estimated in quite different ways, this fact can hardly be taken as significant. What is significant, however, is the type of boy showing considerable differences between Binet I.Q. and performance I.Q.
Taking all the children together, the results from the two scales show the high correlation of .09. Obviously, most of the boys vary very little in their rank order in the two groups of abilities tested. But in twelve cases the variation between I.Q. in Performance and I.Q. in Binet is enough to be judged important (a difference of at least 10 in each case). Considering these briefly in turn : seven have a Binet I.Q. at least 10 higher than their Performance I.Q. What are they like? Brief notes will show.
A. Appears fairly intelligent and is able to learn under close supervision, but is extraordinarily inattentive and restless, unstable and irresponsible. There is practically no hope of social competence in the future. The boy is almost entirely self-centred. His Porteus Maze result?a mental age of five and a half years at the actual age of thirteen?is suggestive.
(Binet I.Q. 71, Performance I.Q. 47.) B. Very talkative, with good language and memory for events; interested and keen on all happenings; but erratic and unstable, spoiling work by over- impetuosity and lack of foresight. (B. I.Q. 66, Perf. I.Q. 49.) C. Can learn, but does not do so with ordinary teaching. Restless, talkative, easily distracted. Shows little power of self-control. (B. I.Q. 67, Perf. I.Q. 57.) D. Amazingly slow in doing anything new. Stubborn; capable of sitting still for hours before assimilating a new idea, and making no effort when work is difficult, although it may actually be within his capacity. (B. I.Q. 61, Perf. I.Q. 44.) E. Nervous to the degree of inhibition of activity when presented with a new task, except when given very much time and encouragement. Can do good routine work, but thought processes seem paralysed by fear. (B. I.Q. 64, Perf. I.Q. 43.) F. Another nervous case. Sits before new work as if inhibited from effort. (B. I.Q. 63, Perf. I.Q. 48.) G. Hardworking and earnest; most anxious to do well; but limited in power to routine and memory work. Very stupid in all practical affairs. (B. I.Q. 56, Perf. I.Q. 44.) t A Study of Performance Tests, by Frances Gaw. 92 MENTAL WELFARE. Five boys have a Performance I.Q. at least 10 higher than their Binet I.Q. Of these? H. Is partially deaf, and has in consequence marked language trouble, which prevents him from doing himself justice on a Binet examination. (B. I.Q. 54, Perf. I.Q. 83.) I. Has a marked physical disability which has probably much interfered with past education. He is slow, but has none of the marked mental characteristic suggestive of permanent defect. (B. I.Q. 93, Perf. I.Q. 108.) J. A borderline case. Difficult to manage, but capable of good work along practical lines. Will probably be socially competent. (B. I.Q. 77, Perf. I.Q. 93.) K. A rather restless, talkative boy, particularly good at all forms of craft work in which he is really interested. (B. I.Q. 65, Perf. I.Q. 79.) L. Another case in which difficulty with many forms of school work is coupled with almost normal intelligence. (B. I.Q. 88, Perf. I.Q. 99.)
Reviewing these cases we see that those showing marked superiority on the Binet scale are of three types, viz. :? (a) Slow, nervous, inhibited children who are afraid to tackle new problems ; (b) Children almost the opposite of these, restless, readily distracted, un- stable ; possessing far more intelligence than they can use; and (c) Children of a slogging temperament, whose patience in acquiring know- ledge gives them a false appearance of power.
With all of these social adaptability and industrial efficiency .are probably far better indicated by their Performance than by their Binet intelligence quotients. With regard to the other cases, the value of the performance tests where special mental or physical disabilities exist is well illustrated. The chance of selecting individuals on a basis of industrial efficiency rather than on aptitude for school work is also shown.
A further examination of the correlations between the results of the Binet tests and those of the Performance scale’ supports these same suggestions. Although the correlation given by the whole set of results is .69, that given if work from three of the boys only is not included is .80. Of these three, one has a marked defect of hearing, and the other two are peculiarly unstable though high grade, and are therefore incapable of much in the way of achievement. Hence it is clear that these characteristics in particular are discriminated by the perform- ance tests.
We may therefore assume that performance tests can be of considerable help in the diagnosis of defect, not only where specific disabilities exist, but also in judging of the chances of industrial and social efficiency in borderline cases. There is, too, reason to hope that increased work on the tests and scales them- selves will add to their possible services in this respect. At present marking the tests, notably estimating the value of the time factor, often constitutes a serious difficulty, more especially when tests tried out with normal children are applied directly to the abnormal. But further research will remedy this.
One other value, by no means the least t:o the educator, of performance tests remains to be considered. So far we have dwelt on their possible uses in diagnosis and in prognosis, and in so doing results have of necessity occupied the focus of attention. To work out mental ages and intelligence quotients results are all that matter : actual achievement alone is under review. But to the educator what a child actually can do at any particular moment is of very minor importance. The things which the teacher must know are what are the child’s difficulties in learning; how the child can and will approach new work with the best hope of success; in other words what the child as an individual really is like and what are the driving forces of his nature. And for successful teaching the sooner this is known the better. Most vitally is it important with candidates for the special school, for they have, as a rule, suffered only too long from the absence of that sympathetic comprehension which alone will prevent in them the establishment of habits of failure so disastrous in their consequences. There can be no doubt that the first stage in the education of the defective is to help him towards some real achievement, to put him in the way of conquering difficulty rather than in the way of being contentedly or sullenly overcome. The more, then, and the more rapidly, one learns about each individual’s nature the better.
For this study few things offer such useful opportunities as the observation of the child during the conduct of performance tests. The casual and undirected observation of the classroom is here replaced by study under definite and fixed conditions. The approach made by different children to the same task can be watched and noted. From this point of view what is done matters not at all : how it is done is everything. What motives govern the child’s behaviour? Is he interested at first with rapidly flagging attention or does his interest grow with effort? How far is the teacher’s presence a necessary impetus to work at all? What is the effect of failure? How far are mistakes recognised and avoided, and how far is the work a meaningless routine? These and many similar ques- tions our study can aim at answering, for it is essential to the very nature of the tests that in them the child must work without help and without interference. So he reveals himself. It is, of course, understood that such study cannot be undertaken in the classroom. The individual child must have during it the observer’s undivided and unflagging attention.
Of its ultimate value there can be no doubt. A few definite records of what a child actually does, of what he achieves and how he behaves under fixed con- ditions, is worth more than any number of vague impressions ; even if the vague impressions be those of an experienced observer. For experience in dealing with and judging children, necessary and irreplacable as it is, is apt to develop habits and prejudices of mind which may seriously lead one astray. What is one to think, for instance, of the character of a bov judged always on ordinary behaviour as reckless, impulsive, unheeding and careless, who sits before every performance test silent and thoughtful until he has arrived at a possible method of solution ? Or of a boy of twelve years, good at school work and judged but slightly defective, who reveals a total inability to control reckless impulse or to plan? In the first case realisation of the possibility of success hinders impetuosity and develops thought; in the second, intelligence is overwhelmed and swamped.
In conclusion, let us summarise briefly our main points. Performance tests are of undoubted service in the study of mental deficiency for two chief ends. Used with cautious discrimination as a scale rather than as individual tests, and with a full realisation of their imperfections, they may act successfully in supple- menting the Binet scale for diagnosis and prognosis. Used as a means of individual study, they can, by providing a fixed method and so giving standards for comparison, be of infinite use to the teacher who seeks to have an intimate knowledge of his individual children so as to help them best.
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.