From the mid-Kloc-0/9th century to the 1980s, the first period of educational evaluation-psychological test period, the research on educational measurement has made a series of achievements, and important progress has been made in the quantification, objectification and standardization of tests. The key point is to use quantitative methods to measure students' learning situation. However, the examinations and tests at that time only required students to memorize the knowledge content of textbooks, which was one-sided. Can not truly reflect the learning process of students. From 1930s to 1950s, it was the second period of educational measurement-the goal-centered period. Taylor (R) put forward the principle of educational evaluation, that is, the Taylor principle of educational evaluation, and clearly put forward the concept of educational evaluation, thus distinguishing educational evaluation from educational measurement. Educational evaluation was born and developed on the basis of Taylor's principle. In the west, people generally call Taylor the father of educational evaluation. The 1960s was the third period of educational measurement ── the standard development period (1950s-1970s). Bloom-based educators put forward the problem of evaluating educational goals, and American educators such as Screvane (M.), Stark (R.E.) and Kellogg (T.E.) evaluated education. Scholars define 1967 as a turning point in the development of American educational evaluation. After the 1970s, educational evaluation developed into the fourth period-the period of results recognition. In the meantime, we are very concerned about the appraisal of the evaluation results. Pay attention to the evaluation process and emphasize that the evaluation in the evaluation process gives individuals more possibility of being recognized. In short, it attaches importance to the constructive role of evaluation in individual development, so it is also called personalized evaluation period.

Modern education evaluation that has appeared since 1930s has the following characteristics:

1. Evaluate the change of purpose. The purpose of early education evaluation is to "choose children suitable for education", while the purpose of modern education evaluation pays more attention to "creating education suitable for children", that is, from attaching importance to identification to paying more attention to the improvement of teaching and learning, so as to form educational goals to the maximum extent;

2. The expansion of the evaluation object. The object of early education evaluation is mainly the teaching field, focusing on the evaluation of students' academic performance, and on this basis, judging the advantages and disadvantages of teaching plan and curriculum development. The object of modern education evaluation extends to all educational fields, and all aspects of macro and micro educational activities can be used as evaluation objects;

3. Innovation in the form of evaluation results. Early education evaluation not only attaches importance to expressing the evaluation results in the form of quantity, but also attaches importance to expressing the evaluation results in the form of language description, and also attaches importance to expressing the evaluation results in the form of combining quantity and description;

4. Emphasize the participation of the evaluation object. In early education evaluation, the evaluation object is regarded as only passively accepting the evaluation, while in modern education evaluation, the evaluation object is regarded as the main body of the evaluation, emphasizing the importance of the self-evaluation of the evaluation object, guiding the evaluation object to participate in the whole process from specifying the evaluation scheme to obtaining the evaluation result, and emphasizing the recognition of the evaluation result by the evaluation object.

There are many different views on the historical development stage of educational evaluation. Levy, editor-in-chief of Encyclopedia of International Courses, made a long-span investigation on the history of educational evaluation, which was divided into three periods: the period of classical examination; The period when psychological measurement is dominant; Postmodern period. Cuba and Lincoln, American evaluation experts, published a monograph entitled "The Fourth Generation Education Evaluation" in 1984, and put forward the basic theoretical viewpoints and framework of the fourth generation education evaluation, which caused great repercussions in the United States. Before this theory was put forward, educational evaluation experienced three theoretical forms, which they called "the first three generations of educational evaluation theory". Its classification standard also basically represents the development course of educational evaluation. The first evaluation theory prevailed in the testing period from the end of 19 to the 1930s, and the test results only reflected students' memory characteristics of knowledge. The second generation evaluation theory appeared with the "eight-year research" in 1930s, represented by Taylor model, characterized by describing the consistency between educational results and educational goals, and it was an evaluation era marked by "description". The third generation evaluation theory originated from the educational reform after 1957 and continued until the 1980s. Its characteristics are not limited to description, but also include judging the advantages and values of education and teaching methods. By the fourth generation of evaluation theory, according to the understanding of Cuba and Lincoln, the shortcomings of the first three generations of evaluation theory are: first, the evaluation object and all other related people are often excluded, which is easy to form a tense and antagonistic relationship between the evaluator and the evaluation object; Second, it ignores the role of other value systems in evaluation. The outstanding problem is the "value difference" caused by culture, and the evaluation adopts "scientific method", which leads to insufficient attention to background factors, lack of flexibility in the evaluation process and neglect of qualitative evaluation methods. In view of the deficiency of traditional evaluation theory, the fourth generation evaluation theory pays more attention to the coordination of the differences of different value systems in evaluation, and regards it as the key issue of evaluation work. More importantly, they believe that the evaluation results do not depend on the same degree of objective reality, but on the consistency of opinions of all participants. People think that the fourth generation evaluation theory will be more practical than any previous generation.