CHAPTER XV

STANDARDIZATION

TESTS AND MEASUREMENTS OF PRODUCTS

There is a group of recent studies which affect the curriculum and all other phases of school organization so profoundly that a separate chapter must be devoted to an exposition of their character and aims. These are studies which aim to standardize school work through tests, measurements, and exact quantitative descriptions of the products of teaching.

For example, one of the efforts of the elementary school is to teach pupils to write. It is entirely possible after the school has done its work to find out by an examination of the results how well pupils can write. It is never expected that pupils in the second grade will write as well as pupils in the upper grades. In this sense, then, it may be said that the results expected in the second grade are of a lower standard than those expected higher in the school. Furthermore, there is a sharp contrast between rapid writing and slow writing. The pupil who writes one hundred and fifty letters in a minute with a quality or form of letters which is fair exhibits one kind of result, while the pupil who writes only seventy-five letters a minute but shows great regularity in his letters exhibits another kind of result. It is not easy in two cases such as have just been described to determine at once which result is better. It may be that speed should be encouraged in order to secure free, fluent movements even if quality has to be sacrificed for the time being.

EARLIER STANDARDS BASED ON OPINION

One difficulty in dealing with the results of school work has been that schools have had no clear definition of what ought to be demanded. Opinion has been matched against opinion. Thus the parent often feels that he has a right to pass unqualified judgment on the progress of his child and on the teacher’s methods of dealing with him. The employer demands of the boy whom he employs a certain proficiency in spelling and adding. The superintendent, in pursuance of his duties, tells the teachers that their work is satisfactory or otherwise and that the children do or do not read as well as they should. The teacher has a certain expectation, and the pupil feels sure that he is doing his work well. Each, according to his personal standard, is estimating the work done in the school.

Very often these standards differ when applied to one and the same performance; sometimes they differ so radically that social troubles follow. The parent says that his child is doing satisfactory work, while the teacher estimates the work as inferior. In such a case it happens, often after a controversy, that one standard ultimately prevails. It is a matter of record in some communities that the parent’s standard has at times been asserted with enough energy to result in the removal of the dissenting teacher from office. On the other hand, it is more commonly true that the teacher’s standard dominates, and the pupil either changes his ways or fails of promotion. In either case, it would have been better for all concerned if some exact standard could have been set up which would have been recognized as superior in its sanction to individual opinion.

Even teachers of experience disagree in grading the same examination paper. One demands correctness in every detail, while the other concentrates attention on originality and force of expression.

OBJECTIVE AND EXACT STANDARDS

The effort to lay down by investigation satisfactory standards of school work is one of the most productive lines of educational inquiry which has ever been instituted. Like all great movements, this movement of standardization has been misunderstood and opposed, but it is steadily gaining ground and promises to be the largest contribution of this generation to education.

In essence it consists of a careful, systematic measurement of what pupils accomplish. If there are at hand measurements of the actual achievements of pupils in various subjects in all the grades, it is safe to compare any single performance with the general average. It should be noted that this does not imply a demand that every pupil’s work be like the average. There are pupils who do their work under unfavorable conditions, as, for example, pupils who have difficulty in reading because they hear no English at home. Their results should not be expected to reach the average, at least in the early grades. How far the results are from the average should, however, be definitely known. Explanation can then be given. Where conditions are not unfavorable the demand can be the more vigorously made that the average expectation be reached.

BEGINNINGS OF THE MOVEMENT

The way in which this movement began and the rapidity with which it has progressed are vividly described by one of its chief exponents as follows:

Eighteen years ago the school superintendents of America, assembled in convention in Indianapolis, discussed the problems then foremost in educational thought and action. At that meeting a distinguished educator[69]—the pioneer and pathfinder among the scientific students of education in America—brought up for discussion the results of his investigations of spelling among the children in the school systems of nineteen cities. These results showed that, taken all in all, the children who spent forty minutes a day for eight years in studying spelling did not spell any better than the children in the schools of other cities where they devoted only ten minutes per day to the study.

The presentation of these data threw that assemblage into consternation, dismay, and indignant protest. But the resulting storm of vigorously voiced opposition was directed not against the methods and results of the investigation, but against the investigator who had pretended to measure the results of teaching spelling by testing the ability of children to spell.

In terms of scathing denunciation the educators there present and the pedagogical experts, who reported the deliberations of the meeting in the educational press, characterized as silly, dangerous, and from every viewpoint reprehensible, the attempt to test the efficiency of the teacher by finding out what the pupils could do. With striking unanimity they voiced the conviction that any attempt to evaluate the teaching of spelling in terms of the ability of the pupils to spell was essentially impossible and based on a profound misconception of the function of education.

Last month in the city of Cincinnati that same association of school superintendents, again assembled in convention, devoted fifty-seven addresses and discussions to tests and measurements of educational efficiency. The basal proposition underlying this entire mass of discussion was that the effectiveness of the school, the methods, and the teachers must be measured in terms of the results secured.[70]

HANDWRITING SCALES

One of the earliest types of school work to be standardized was handwriting. Standard “scales,” as they are called, have been prepared by several investigators, and their use has become very common.

The first scale was prepared by Professor Thorndike.[71] He secured a number of specimens of children’s writing, and asked experienced judges to arrange these specimens in a series of descending degrees of excellence. By combining the judgments returned by the experts it was possible to secure an average judgment. Certain typical specimens were then set aside, representing equal steps in the descending scale. In practical use, a given sample of handwriting which is to be judged is compared with the successive steps in the scale until an approximate equality in degree of excellence is found. The sample to be judged is then marked with the grade agreed on for the standard specimen.

A second scale was prepared by Ayres[72] on a more objective basis. The specimens were arranged in a series, not in accordance with the judgment of experts, but according to the time which was required to read them.

A third scale, more elaborate than either of the others, was prepared by Freeman.[73] He first made an analysis of the different characteristics which enter into excellent writing, such as uniformity of slant, uniformity in the height and spacing of letters, and other like essential characteristics, and then selected specimens exhibiting decreasing grades of excellence in each of these characteristics. Since each characteristic of writing is capable of definite measurement, the specimens could be graded on the basis of direct measurements. Thus the slant of a number of specimens was measured letter by letter, and objective grades were established.

Finally, the preparation of scales of handwriting has gone so far that special scales or series of graded specimens for particular school systems have been prepared.

SPEED AS A CORRELATE OF QUALITY

In the meantime the matter of speed in handwriting has also been a subject of careful measurement, and tables of average speeds for different grades have been prepared in a number of school systems.

A device for presenting in a single diagram both speed and quality and at the same time comparing several grades in the same school with each other was worked out in the Cleveland survey. The figure and a description of its meaning are given on pages 218, 219.

The relative emphasis on speed and quality actually found in a number of different schools is set forth in the following diagram. The separate parts of this diagram are made up as follows: The average speed of a grade is represented by distances in the horizontal, and average quality by distances in the vertical, scale. Thus, taking the first section of the diagram, that of the North Doan School, the fifth grade has an average speed of 71 letters per minute, and an average quality of 41. The sixth grade shows progress in both speed and quality, though speed increases more than quality. The seventh and eighth grades show further progress in both speed and quality, the two changing at about the same rate. The diagram for the Kentucky School shows progress of a slightly different type. In this school the sixth grade, as compared with the fifth, shows progress in quality, but very little in speed. Progress from the sixth grade on is about equal in quality and speed. Memorial School emphasizes speed almost exclusively up to the eighth grade, while Mt. Pleasant emphasizes quality.

The various schools which have been reported in the four upper sections of the diagram are all regular in the sense that each school shows steady progress from grade to grade in both speed and quality. Without attempting to comment in detail on the special cases, attention is called to the series of results presented in the lower part of the diagram.[74]

Before giving examples of standardization in fields other than penmanship, it will be well to indicate the full meaning of the foregoing paragraphs.

STANDARDS, PERSONAL AND IMPERSONAL

First, it will be seen that measurements are here substituted for purely personal judgments. It was the universal practice before this movement began, and it is the common practice to-day, for a supervisor to go from school to school, passing on the excellence and speed of handwriting. The supervisor has arrived at a personal standard through his experience. He expects a certain result in the fifth grade. He has in his mind a more or less clearly defined requirement and regards it as his duty to impose this on pupils and teachers. In the same way the teacher has a personal standard which is imposed on the pupils. It would be a mistake to describe these personal standards as arbitrary or unintelligent. The experienced teacher is usually approximately right in his expectations, and the supervisor usually does a great deal to raise and unify the level of work with which he comes in contact. But it is not possible in a complex social situation to rely on personal standards. Personal life and professional activity are too transient. How many schools have changed standards, to the great disadvantage of the penmanship, with each change of supervisors? Furthermore, personal standards are vague when one tries to transmit them to others. This is a serious matter when it is remembered that a very large proportion of teachers in each school are changing each year. If a supervisor is to systematize the work of his schools, he must constantly be bringing into agreement with his standards the standards of a large number of new teachers. For the sake of coöperation it is advantageous to turn a personal standard into one which can be described and defined. Finally, a personal standard grows up out of all the accidents of a personal career. The person of narrow experience may not have incorporated into his standards valuable elements which he would have accepted had he come in contact with them. The person of strong personal likes and dislikes may often be prejudiced. The person of broad experience may be inexact when it comes to details. To demand of all who teach handwriting that they rise above purely personal standards is not unlike the demand that the central government rather than the states mint our coins.

SOCIAL STANDARDS VERSUS IMPOSED STANDARDS

Second, the standards set up are derived from the work actually going on in schools. There is no dictation from purely theoretical and arbitrary sources. It is quite impossible to close one’s eyes to the fact that in the past there has been a tendency to assume that the only standard of action is the perfect standard. Many a child has been taught penmanship from perfect copy and has been urged to imitate this copy at whatever cost of time and pains. The slow, painful effort to draw letters like those in the copy book is not an unfamiliar exhibition in the penmanship class. A standard derived from the school work itself is a social standard; it is based on what pupils really do. One need not be satisfied with present performances, but one starts from solid ground. Furthermore, out of actual measurements will come a clear idea of the range of variation. One of the most astonishing facts which have come out in the course of the study of standards is the fact that there are very wide variations in the same grade. There is, therefore, an easy possibility of finding for each grade high standards. These high standards have the further advantage of being standards actually realized by pupils. We are justified in describing the standards thus set up as natural standards. They do not limit the progress of any grade or aim at mechanical uniformity as do the arbitrary standards based on personal judgments.

COMPARISON THROUGH EXACT MEASUREMENT

Third, standards measured and expressed in definite terms can be compared and can be made the basis of studies which are quite impossible so long as standards are not expressed in common terms. For example, when the speed of handwriting in a certain school is deliberately changed, what is the effect produced on quality? Heretofore it has been almost impossible to answer such a question. Every school reform has been enthusiastically hailed by its friends as accomplishing much. In the second generation most of these reforms are checked, if not actually dropped, because it is found that the good accomplished in one line is entirely lost in some other. To-day reforms are in a position to measure their effects in all directions. A change in the speed of handwriting may or may not be advantageous; it is the duty of measurement to so state results that some light will be thrown on this matter.

A concrete example will serve to show how studies of this kind may be carried out. Fig. 15 shows the relative speeds and qualities of handwriting found in various grades in a miscellaneous group of cities and the corresponding facts for St. Louis and Grand Rapids. It is seen that both of these school systems are ahead at all points in speed and behind at first in quality. Both cities have made a conscious effort to get away from the slow drawing of letters in the lower grades. In doing this quality has been sacrificed. It is not the purpose of this discussion to decide what methods of teaching handwriting are best; the value of this example is that it shows quality and speed reduced to terms where the two can be studied together and with a high degree of exactness.

RECORDS AS A BASIS OF STANDARDIZATION

Fourth and finally, measured standards show the direction in which pupils are moving, because they permit a permanent record of each step of the child’s development. Schools have been slow to learn the value of records. On the one hand, school records have been piled up by the tome and no use has been made of them; on the other hand, they are usually so loosely thrown together that they are of very little value in guiding educational policy. Here is a form of record which can be duplicated and compared from year to year. Medicine has long since learned that exact records are the only safe means of guiding treatment. Modern agriculture has become scientific through the use of records and through decisions regarding experiments which these records make possible. Modern business has learned to make its accounting intelligent enough to guide policies. Finally, schools are beginning to see that records of a type permitting continuous comparisons are invaluable in determining at what point school work shall take this or that form.

STUDIES OF ORAL READING

What has been done with penmanship has been paralleled in some other subjects of elementary instruction. The following quotations have to do with oral reading:

A coöperative study of reading was organized during the month of September by the committee in charge of the grade-teachers’ section of the Illinois State Teachers Association (Northeastern Section), which met at Elgin, Illinois, November 3 and 4. The purpose of this study was to secure a body of facts in regard to the achievement of boys and girls in reading in a number of schools represented in the Association....

The materials used in this study of reading were the standardized oral-reading paragraphs and the silent-reading tests which have been used in connection with the surveys in Cleveland, Grand Rapids, and St. Louis, as well as in a large number of investigations carried on in other cities....

The standardized oral-reading paragraphs consist of a series of twelve paragraphs arranged in the order of increasing difficulty. The tests were given to the pupils individually by a principal or by a teacher who had been previously trained for the work. As the pupil read the teacher recorded the time required to read each paragraph together with the number of errors which were made of the following types:

(_a_) Gross mispronunciations, which include such errors in pronunciation as indicate clearly that the word is too difficult for the pupil to pronounce.

(_b_) Minor mispronunciations, which include the pronunciation of a portion of a word, wrong accent, wrong syllabification, omission of syllables, etc.

(_c_) Omission of words.

(_d_) Insertion of words.

(_e_) Repetition of words or groups of words.

(_f_) Substitution of one word or group of words for another.

A pupil continued to read until he had made seven or more errors in each of two paragraphs. By means of a system of scoring based on the time required to read and on the number of errors which were made it was possible to represent the achievement of a pupil or a class in numerical terms....

The upper section of the table [given below] gives the average number of seconds required to read paragraph 1 and the average number of errors made by three _poor_ second-grade classes and by three _good_ second-grade classes. Of the poor schools, School M made more errors and read more slowly than the average. School N read with fewer errors than the average, but read so slowly that the oral-reading score for the class was below the average. School O, on the other hand, gave sufficient emphasis to rate, but neglected accuracy to such an extent that the oral-reading score was low. An examination of the records made by the good schools shows clearly that consistent progress in both rate and accuracy is a prerequisite to a high level of achievement. The schools of northern Illinois vary widely in the amount of emphasis given to these phases of reading achievement. There is need, on the part of many teachers, for a continuous critical study of the specific character of the results which they are securing.

RATE AND ERRORS IN ORAL READING

PARAGRAPH 1—GRADE II

===================================================================== | | POOR SCHOOLS | GOOD SCHOOLS |AVER-|—————————————————+—————————————————- | AGE | M | N | O | X | Y | Z ——————————————————————————+—————+—————+—————+—————+—————+————-+—————— Rate [seconds per passage]| 42.2| 65.0| 64.1| 39.1| 27.2| 32.8| 37.9 Errors | 1.4| 2.0| 0.5| 2.4| 1.1| 0.9| 1.3 =====================================================================

PARAGRAPH 1—GRADES II, III, AND IV

=============================================================== | AVERAGE | SCHOOL A | SCHOOL B | SCHOOL C | SCHOOL D ——————————+—————————+——————————+——————————+——————————+————————— Grade II | | | | | Rate | 42.2 | 37.9 | 65.0 | 39.1 | 43.4 Errors | 1.4 | 1.3 | 2.0 | 2.4 | 1.7 Grade III | | | | | Rate | 21.9 | 19.8 | 23.6 | 23.9 | 28.0 Errors | 0.9 | 0.7 | 1.7 | 1.8 | 0.8 Grade IV | | | | | Rate | 18.6 | 18.0 | 21.9 | 16.0 | 27.0 Errors | 0.8 | 0.6 | 1.3 | 0.5 | 1.5 ===============================================================

Additional light is thrown on this problem when we follow certain schools through the second, third, and fourth grades. The average rate and number of errors for Grades II, III, and IV are given in the left-hand column of the lower section of the table. The records for School A show that second-grade pupils do better both in rate and in accuracy than the average. The same thing may be said of the third and fourth grades. Continuous, consistent progress of this type is very commendable. In School B, on the other hand, the pupils do less well in each grade in both speed and accuracy than the average. A question arises here concerning the general effectiveness of the classroom instruction. School C ranks low in accuracy in the second grade. Apparently this difficulty was realized in the third grade, and considerable progress both in speed and in accuracy resulted. In the fourth grade average results are attained which are above the average. This school represents consistent, continuous growth from grade to grade of a highly desirable type. School D, on the other hand, makes improvement in speed and accuracy in the third grade, but fails to increase its rate in the fourth grade, and makes a record in accuracy which is distinctly below the record made by the third grade. It is evident, if the records for the present second, third, and fourth grades are typical of the results secured from year to year, that there is need for more intelligent instruction and supervision in School D.

In this connection it should be said that objective standards of attainment for each grade should be defined. By means of tests given throughout a school or a city the present level of achievement can be determined. By means of comparisons with results secured elsewhere new goals of attainment can be defined. Each teacher should become familiar with the methods of giving tests. She should utilize them frequently in examining her work to find sources of strength and weakness. Through the co-operation of teachers and supervisors progressive revisions in standards of attainment and methods of procedure should be made. This type of co-operation is necessary because it is only when all the units of a school system work consistently together toward clearly defined ends that the most effective results can be secured.[76]

STUDIES DEALING WITH OTHER SUBJECTS

A great number of similar studies are being reported each year on arithmetic, spelling, and other aspects of the elementary curriculum. The high-school subjects are more complicated than those in the elementary school, but even these are beginning to be tested. There are satisfactory tests in algebra and the beginnings of measurements in Latin and English.

MECHANICAL ASPECTS THE FIRST TO BE STANDARDIZED

In all cases standardization begins with the mechanical aspects of school work. These are more susceptible to exact quantitative description and are the first to be taken up. Some writers have professed to find in this a reason for rejecting the whole movement toward standardization. There are, they assert, products of teaching which are subtle and intangible. These are the products which are most highly to be prized. Thoroughly to standardize penmanship and oral reading and algebra is to set aside these more important matters.

Two answers are to be made to this objection to the movement toward standardization. In the first place, the higher values of education are not secured by teachers who are negligent of the fundamental mechanical requirements. The teacher who successfully trains his pupil to study history will make of him a good reader also. In the second place, if it should prove to be desirable to give less time than is given at present to training in the mechanical aspects of school subjects, it will certainly be absolutely essential that the limits and restrictions be set up with discrimination. We shall never be able to deal intelligently with the mechanical aspects of education until we have studied them.

A third statement which can be ventured with assurance in the light of the recent history of this movement is that its limits cannot be set. Each year new aspects of school work are measured with exactness. It is certain that the ultimate conquests of measurement will push the opponents back into their own territory.

STANDARDIZATION AND THE SCIENCE OF EDUCATION

In short, standardization is nothing but a systematic effort to deal with educational problems explicitly and in the light of exact information. Whatever may be the limits of exact knowledge in educational matters, it is certain that we ought to secure as much knowledge of this type as possible.

EXERCISES AND READINGS

The exercise which will best serve to supplement this chapter is a series of tests performed on members of the class and worked out by them for purposes of comparison with other standard results. In the appendix of the volume of the Cleveland survey entitled “Measuring the Work of the Public Schools” a full set of standard tests will be found.

S. A. Courtis, 82 Eliot Street, Detroit, Michigan, furnishes tests in various subjects, especially arithmetic.

The following institutions furnish various tests:

College of Education, The University of Chicago, Chicago, Illinois.

Teachers College, Columbia University, New York City.

Bureau of Measurements and Tests, State Normal School, Emporia, Kansas.

The readings which are most useful in this connection are to be found in current educational periodicals. The student will find the latest scientific studies in such journals as the following:

_School Review._ Published by The University of Chicago Press. This is a journal dealing chiefly with high schools.

_Elementary School Journal._ Published by The University of Chicago Press. This contains very full reviews of elementary tests.

_Journal of Educational Psychology._ Published by Warwick and York, Baltimore, Maryland.

_Educational Administration and Supervision._ Published by Warwick and York, Baltimore, Maryland.

_Educational Review._ Published by the Educational Review Publishing Company, Easton, Pennsylvania.

_School and Society._ Published by The Science Press, New York City.

FOOTNOTES:

[69] J. M. Rice, editor of the _Forum_.

[70] Leonard P. Ayres, “Making Education Definite.” _Bulletin No. 11_, Indiana University, Vol. XIII (October, 1915), pp. 85-86. Published by the Extension Division of Indiana University.

[71] E. L. Thorndike, “Handwriting.” _Teachers College Record_, March, 1910.

[72] L. P. Ayres, A Scale for Measuring the Quality of Handwriting of School Children. Russell Sage Foundation, New York.

[73] F. N. Freeman, The Teaching of Handwriting. Houghton Mifflin Company, 1914.

[74] Measuring the Work of the Public Schools, pp. 75-77. Cleveland Education Survey. Published by the Survey Committee of the Cleveland Foundation, Cleveland, Ohio, 1916.

[75] Quality on vertical scale, speed on horizontal scale. The four schools referred to in the text are represented in the four diagrams in the upper part of the figure. North Doan is reported in the diagram in the upper left-hand corner. Kentucky is shown in the upper right-hand corner. Memorial is under North Doan. Mt. Pleasant is under Kentucky.

[76] William S. Gray, “A Co-operative Study of Reading in Eleven Cities of Northern Illinois.” _Elementary School Journal_, Vol. XVII, No. 4 (December, 1916), pp. 250-257.