In this 1941 Kappan article, G.T. Buswell sums up the status of education research and proposes new directions for the future.
There are many evidences that the vigor of the educational research movement during the two decades from 1910 to 1930 has not been equaled by the efforts of the last ten years. There have been fewer important techniques developed and much of recent research can be characterized as an extension of methods of study developed prior to 1930 and lacking in a high degree of fertility in sensing new methods for attacking problems of education. Educational research seems to be on a plateau.
There appears to be considerable concern over the present status of research. Only a few years ago the National Society for the Study of Education devoted an entire yearbook to The Scientific Movement in Education. More recently a committee of the American Council on Education has been studying the nature of educational research and has produced some initial publications dealing with this problem. In September of the present year the University of Chicago devoted one of its symposia on the occasion of its Fiftieth Anniversary Celebration to the problem of the conceptual structure of educational research. On that program three stimulating papers were presented by Professors T. R. McConnell, Douglas E. Scates, and Frank N. Freeman. These and other publications relating to research indicate that the scientific study of education is outgrowing its infancy and has now matured to the point where it is ready for a higher level of expression.
A careful review of the progress of educational research during the present century does not lead the writer to any feeling of discouragement as to the possible contribution of a scientific study of educational problems. Quite to the contrary, it leads to the belief that the achievements in research up to the present time have contributed mightily to a vigorous development of public education and that they challenge the present generation of students to develop a structure of educational research on a higher plane than was possible when the movement was new.
A brief overview of the trends in research since 1900 will lay a basis for proposals for its present improvement. Illustrations will be drawn from four fields of educational research.
The psychological study of mental abilities had little to build upon in the year 1900 except some of the initial studies of Cattell and his students made during the 1890’s. Some of the present concepts for measuring mental ability had not yet been developed. It was not until 1908 that the first edition of the Binet-Simon test was made available and not until 1916 that the Stanford Revision by Terman was published. During the early stages of the intelligence testing movement, experiments were carried on with many miscellaneous tests. Some of these resembled ordinary parlor tricks and research with them could often be described as little more than tinkering. However, with the emergence of certain concepts, the structure of research in mental measurements began to take definite form. The concept of mental age was one of these and, later on, the concept of a ratio measurement, such as an I.Q., gave much flexibility to research in this field. As the measurement movement expanded and began to produce achievement tests as well as intelligence tests, important statistical concepts were developed. The publication of Thorndike’s Mental and Social Measurements as well as other books on statistical methods contributed much toward the refinement of the measuring scales that were being built. The measurement movement, surveyed as a whole, ·furnishes an excellent illustration of how one field of research grew from an immature tinkering with mental abilities to a rather highly developed method of measuring. We frequently deplore what is sometimes described as a tinkering variety of research, and yet even this tinkering proved to be far more productive than the a priori arguing which characterized the handling of education problems prior to the application of research techniques to their solution.
Another example of the development of research might be drawn from the field of learning and the application of laboratory techniques to the study of such problems. With the introduction of objective laboratory techniques by Wundt in 1878, a new impetus was given to the study of problems of learning. Judd, who received his training from Wundt, brought to America some of these new techniques plus a high degree of fertility in sensing their applications to problems of learning in school. His early publication of Genetic Psychology for Teachers started a line of research which has been immensely productive in influencing both methodology and content in many fields of learning. Here again some of the early studies seem to be little more than tinkering with laboratory apparatus, but as the techniques developed, new directing concepts were formulated and at the present time this type of research has reached a considerable degree of maturity in several fields of education.
In the field of the curriculum the techniques of educational research were at first of the simplest kinds, the most frequently used being nothing more than counting. Such contributions as Ayres’ tabulation of the thousand commonest words in the English language, followed by other more extensive studies, have provided a ground work for a spelling curriculum which would have been impossible to attain by the older methods of intuition as to what words a child should learn to spell. This notion of frequency-count was then applied to other curricular fields, sometimes productively and sometimes not. The development of a theory of social utility in curriculum construction depended directly on researches in counting frequencies. This technique as applied to the curriculum reached its peak in some of the very extensive studies of Charters and his students. These studies have waned somewhat due to the failure of their proponents to develop any adequate theory for interpreting the frequencies which were so painstakingly derived. Nevertheless, the objectivity of the frequency studies has eliminated the easygoing subjective method of determining what the content of a curriculum should be and the findings of these studies may still prove to be productive when students of the curriculum carry their thinking far enough to arrive at a defensible general theory of education.
Researches in personality have followed a different pattern from those in the three fields just enumerated. The early scientific studies in this field, as for example, Miinsterberg’ s work in vocational guidance, were shunted away from applications to education and there has been a considerable gap, both in time and in kind of techniques employed, between the early work in this field and the present interest in personality and guidance. The clean-cut scientific studies in this field are fewer in number than in the fields of mental measurement, learning, and curriculum. Some of the work now being done in personality is carried on in apparent ignorance of some of the basic concepts of research which are commonplace in other fields. Objectivity of data is less emphasized, reliability of findings is often overlooked. Nevertheless, there are evidences of fertility in developing new research techniques which, when they mature, may make significant contributions.
In the four fields just mentioned a certain pattern of development seems to be common. The antecedent of the initial scientific studies in these fields was in each case an a priori reasoning without adequate data for a basis. These pedagogical discussions resulted in a widespread reaction against the futility of talking without evidence. The early research studies attempted to produce some objective facts which could form a basis for dealing with educational problems. Necessarily the early techniques were simple and in many cases went little beyond ordinary counting. Gradually new concepts for refining research were introduced. Fertility in developing new techniques accounted for a considerable share in the building of a body of basic quantitative data concerning many educational problems. The easy stages of this kind of research have now for the most part been exhausted. New techniques appear less commonly and there is appearing much research, the findings of which are new only to the person who produces them. Educational research is ripe for a more mature set of procedures. These procedures can scarcely be developed until the basic concepts now available are set forth more clearly.
The present writer does not have the temerity to attempt to list or define those concepts which are necessary for building a higher structure of educational research. He will, however, risk mentioning a few examples.
Productive thinking in the field of mental abilities is very clearly affected by the development of certain basic concepts. Distinguishing between concepts of mental capacity and mental functioning has done much to clarify thinking. The concept of an age-grade scale has furnished the technique which has been productive of much valuable work. The concept of primary mental abilities is now influencing research in mental measurements in a way which is likely to prove very productive.
In the field of statistical techniques several concepts are so well known and so widely accepted that no mature student would think ·of carrying on research without adequate attention to them. For example, the concept of a normal distribution curve as contrasted with the older notion of a rectangular distribution of abilities has so influenced thinking that the whole pattern of expected educational achievement is adjusted to this notion. The concept of reliability of a sample is now so clearly established that no competent student would attempt to defend a conclusion regarding a basic issue without careful attention to this aspect of his data. Only the amateur disregards the reliability of the samples which he uses.
In experimental studies the concept of a control group in research procedure is now so well established that it is commonplace among competent students. The structure of experimental research has been widely influenced by this one concept. The concept of a single variable in experimental research is another example of a method of control which has added much to the precision of experimental findings. In nearly every subject field there are some mature concepts which serve to direct and define research in those fields. For example, in the field of reading, thirty years of research on eye movements have served to clarify certain basic concepts relating to the perceptual processes in reading, thereby simplifying remedial techniques in a field which is much afflicted by a superabundance of miscellaneous proposals.
There are undoubtedly some ideas very simple in character but very significant in value which will still be picked up by the amateur. But the field of these ideas has been pretty well picked over and the frequency of such fertility is decreasing in marked fashion as compared with the frequency with which apparently simple ideas were developed into significant techniques and concepts during the two decades preceding 1930. This means that if research in education is to go forward, departments of education must give even more attention to developing in students an understanding of those concepts which are essential to the conduct of research at a mature level.
We are now at the stage where the data in a field must be related to an attempt to develop explanatory theory. In place of tinkering on problems, an attack must be preceded by familiarity with previous work in the field and particularly by an understanding of those concepts which will make possible the formulation of significant hypotheses. Education now is suffering much from a lack of mature hypotheses to guide its research. A new generation in education, discouraged by the delay in pushing scientific work to higher levels of maturity, is resorting to the practices of the pre-research era in education and is attempting to solve its problems by a method of talk, sometimes with complete innocence of a knowledge of the available data relating to its problems. The futility of such talk was what stimulated scientific research in the first place.
In summary, the proposals of this paper are as follows: Educational research has developed through a forty-year period of study an immense body of data, some of which are important, many of which are inconsequential. Furthermore, it has developed a series of research techniques which have proved their values in the search for truth. However, many of these techniques have been at a simple level and the majority of the researches have been carried on without the benefit of adequate hypotheses and without the direction of guiding concepts which have grown out of previous research. This kind of procedure is to be expected in the development of a new field and, in spite of its limitations, has produced results of immense value to education.
However, there are evidences that the productivity of these simple techniques and methods of research is diminishing. The proposal of this paper is that there is now a need for an assessment of the basic concepts produced by research which will guide both in further research and in applications of the findings of previous research to educational problems. This is a task of interpretation and integration and must be done by men who are familiar with research and who sense what clean-cut concepts and clean-cut hypotheses are. It is a job which needs to be prosecuted vigorously and which should result in a series of concepts validated by research which will make it possible to develop a structure for further research in education at markedly higher level than has been possible up to the present time.
The objectives of this higher research should be much more than the development of a body of knowledge. It should be the development of a body of theory which rests upon data interpreted through concepts of the highest order that are now possible. From such an undertaking there should result guiding principles for the conduct of education, principles of such a degree of generality that an understanding of them would cover many specific problems faced by an individual. Research is never complete until it provides a basis for defensible theory.
Citation: Buswell, G.T. (1941). Structure of educational research. Phi Delta Kappan, 24 (4), 167-169, 174.