School Evaluation
学校的评价
The school's evaluation, and the decisions that adopt it, change the roads a child can take; they are a power, and subject to checks. This paper states five propositions. That a child cannot walk away and cannot yet say no both make the effect of this power deeper; where the school adopts automated evaluation, it answers for its adoption. The school may test whether its teaching has reached the children; it does not register meeting some indicator as having replaced judgment of particular work and a particular child. Where arrangements such as public rankings, role models and model classes set up one child as the measure of another child's whole person, the school answers for them. Entrance examinations for the next stage often sit outside the school and are a chokepoint; the school answers for how much of their requirements it brings in, and the party that assesses the school by positions before the gate answers separately for its assessment. Report cards and comments state what was verified and on what basis the evaluation was made, and do not vouch for what kind of person the child is. This paper does not say that tests ought to be abolished, does not say which grading system is best, and does not design entrance examinations.
Writing Declaration: This paper was co-drafted with Claude (Anthropic). All intellectual decisions, framework design, and final editorial judgments were made by the author.
Introduction
In 2021 the General Office of China's Ministry of Education issued a notice on managing examinations, which says: "examination results are not ranked and not made public; students and parents are informed in an appropriate way, and examination results may not be circulated in parents' groups of any kind" (my translation; 考试结果不排名、不公布,以适当方式告知学生和家长,不得将考试结果在各类家长群传播). A rule had reined in rankings first.
The Theory of the University said: "Evaluation is power, and is subject to checks: what an advisor gives a student can be cultivation, and the evaluation in his hands is still power". A university student can change his major, change his advisor, leave. A child in primary school cannot walk away, and in some matters cannot yet say no. This paper asks how deep the effect of the school's evaluation goes when it lands on such a child, and how far the school answers for it.
§1 Evaluation Is Power
It changes the roads one can take. The school's evaluation comes in several kinds: marks in a course, conclusions on promotion and class assignment, end-of-term comments. Decisions that adopt evaluation are included too: assigning classes by one test, recommending by one set of comments. They change the roads a child can take: which class he joins, which teacher he has, whether he can move on to the next stage. Changing the roads another can take, through the asymmetric effect of a position on him, is power.
Checks follow the effect. Political Economy says: "The larger an economic position's power-face, the stronger the checks on it." It also says: "The power-face is compared by its particular effects: whom it converges, down to which dimension, how many must pass through it, how great the cost of walking away". The Theory of the University extended this to the university's evaluation. This paper extends it to children: the school sets checks on its evaluation, and the strength of the checks follows the actual effect. This extension is this series' own undertaking.
Why deeper. This sentence is descriptive. He cannot walk away (Paper 4): a conclusion on class assignment is one he cannot avoid, and he has to spend a year in that class. He cannot yet say no (Paper 3): when he says "that's wrong" about a comment, his words may not land. Both make the effect of evaluation deeper. The deeper the effect, the more the checks matter: grounds that can be stated, a road of review, conclusions not fixed forever.
Automated evaluation. Where the school adopts automated tools of evaluation, allocation or recommendation, it answers for its adoption: what it uses them to evaluate, on what grounds it adopted them, and at which points their conclusions still need a person's judgment. Paper 1 said that tools can teach but do not take the student's step for him; this is a proposition of the same shape. The Theory of the University also said: when evaluation is handed to a machine, the bearing is not handed over with it.
XX26 Evaluation is power, and matters more for those who cannot walk away. The school's evaluation (marks in courses, conclusions on promotion and class assignment, comments), together with decisions that adopt evaluation, changes the roads a child can take; the school sets checks on it, and the strength of the checks follows the actual effect; not being able to walk away and not yet being able to say no both make the effect deeper. Where the school adopts automated evaluation, allocation or recommendation, it answers for its adoption.
§2 What Can Be Measured Is Not All
It can be tested. The school may test whether its teaching has reached the children. Paper 1 said that teaching can be designed and can be done for many at once; whether it has reached them can also be tested. This paper is not against tests.
What can be measured is not all. Education as Subject-Condition says: "Measurability is a necessary condition at the instrumental level. The problem is when it is treated as existence itself and as purpose itself." The school does not register meeting some indicator as having replaced judgment of particular work and a particular child, nor does it use an indicator to exempt the evaluating position from explaining and answering. That a child scored ninety in mathematics shows first that he got ninety under the marking rules of this paper; the total alone is not enough to decide where exactly his understanding has reached, still less what kind of child he is. Where his understanding has reached takes going back over the questions, the answers and the grounds of the marking. The Theory of the University said a proposition of the same shape at the level of the university: evaluation does not turn judgment back into rules.
Indicators get bent. The Theory of the University held a dialogue with Campbell's 1976 observation: the more a quantitative social indicator is used for social decision-making, the greater the pressure to corrupt it. What Campbell put forward is an empirical tendency. This series does not undertake its general strength, does not derive normative judgments from it, and judges only by particular effects. Colorado's provision read in Paper 5, which writes students' academic growth as a statutory share of a teacher's evaluation, serves as a contrast here too.
An objection. One view goes further: schooling leads people to confuse process with substance, to take having attended class for having learned, and having a diploma for being capable; and evaluation is the core of this confusion.1 This paper agrees that what can be measured is not all. But the view reaches beyond evaluation: compulsory schooling, the link between credentials and opportunity, the school's monopoly on learning. What this series can answer is spread over several places: the side of not being able to walk away, Paper 4; the side of credentials becoming a gate, §4 and Paper 7; records being only a limited proof, §5. On the school's monopoly on learning, the Prequel, §5, said that children who learn elsewhere do not lose their eligibility to return to public provision; the answer goes that far. What cannot be answered is left as it stands.
XX27 What can be measured is not all. The school may test whether its teaching has reached the children; it does not register meeting some indicator as having replaced judgment of particular work and a particular child, nor use it to exempt the evaluating position from explaining and answering.
§3 Not Setting Up One Child as the Measure of Another
That sentence in Teacher-Student Theory. Teacher-Student Theory writes: "Where one student is used as the measure of another, for instance 'look at so-and-so,' what is read first is how the sentence lands with the student who hears it." It goes on: "If it also actually alters the situation of the one used as the measure, that relation is entered separately"; and: "Using a piece of homework to show one way of doing something, and setting up one student as the measure of another student's whole person, are read apart."
Extended to the school. The school has such arrangements too, only they do not come from one teacher's sentence: public rankings, honor rolls, role models, model classes. This paper extends the sentence in Teacher-Student Theory to the school's arrangements, and that step is this series' own undertaking. First read how it lands on each child: what the child near the bottom reads off this board. Where one child is set up as the measure of another child's whole person, the school answers for it. Where it also actually alters the situation of the child used as the measure, for instance he is compared in public, or assigned to another class on that basis, that is entered separately. Using a piece of homework to show one way of doing something is not of this kind; that there are rankings is not in itself judged as crossing the line.
A notice. Besides not ranking and not making public, the 2021 notice also says: "mid-term and end-of-term examinations in schools use graded evaluation, generally in 4 to 5 levels" (my translation; 学校期中期末考试实行等级评价,一般分4至5个等级); no pen-and-paper examinations in the first and second grades of primary school; "schools and classes may not organize weekly tests, monthly tests, unit tests or other examinations of any kind" (my translation; 学校和班级不得组织周考、月考、单元考试等其他各类考试).2 The same notice says: "class assignments and seating may not be adjusted, nor students 'labeled,' according to examination results" (my translation; 不得按考试结果给学生调整分班、排座位、“贴标签”); assigning classes by examination results and labeling children have an express provision here. Two sides: not making rankings public is written into the rule; not making them public is not the same as not ranking, and how much was done is checked separately. In 2025, Article 14 of the General Office's notice on managing routine examinations says: "routine examinations in compulsory education schools use graded evaluation; examination results are not ranked and not made public, students and parents are informed in an appropriate way, and the tendency to count scores alone is to be resolutely overcome" (my translation; 义务教育学校日常考试实行等级评价,考试结果不排名、不公布,以适当方式告知学生和家长,坚决克服唯分数的倾向。). This one does not repeat the sentence "generally in 4 to 5 levels."
A contrast. An explanation by Japan's Ministry of Education, Culture, Sports, Science and Technology says that the overall assessment in each subject, too, "has been changed from 'evaluation referenced to the group' (so-called relative evaluation) to 'evaluation referenced to objectives' (so-called absolute evaluation)" (my translation; 「集団に準拠した評価」(いわゆる相対評価)から、「目標に準拠した評価」(いわゆる絶対評価)に改めたところです), that is, from grading by the distribution of classmates to grading by objectives. Grading by distribution is one way a system writes classmates in as the reference for assessment; here it serves only as a contrast of reference frames. Whether it goes further and sets other children up as the measure of a whole person is checked separately by how the school uses the grades; grading by objectives does not by itself prove that the school has not characterized children through comments or class assignment.3
A rule that makes rankings public, to serve as the opposite case, this paper has not yet found.
XX28 Not setting up one child as the measure of another. Arrangements such as public rankings, role models and model classes are read first by how they land on each child; where one child is set up as the measure of another child's whole person, the school answers for it; where it also actually alters the situation of the child used as the measure (compared in public, assigned to a class on that basis), that is entered separately. Using a piece of homework to show one way of doing something is not of this kind; that there are rankings is not in itself judged as crossing the line.
§4 The Gate of Progression
The gate is outside, and also inside. Entrance examinations for the next stage often sit outside the school, and are a chokepoint: a position others cannot get around on the way to unfolding their own roads. Political Economy says: "When qualifications become a door, they are a chokepoint." Where the school's grades, comments and recommendations are adopted by the gate, and in moving up and streaming within the school, in these places the school is inside the gate.
How much is brought in. Education as Subject-Condition says that the years most in need of being kept for children are filled with drilling and grade points; it also says: "'Study hard, get into a good university, find a good job' uses a computable path to impersonate a purpose." This paper does not judge that road at the level of the school. The school answers for how it brings in and adopts the requirements of the chokepoint of progression: how much time preparation for the gate takes, and what it squeezes out; whether teaching, care, response to particular questions, and time with no preset use remain; whether students who do not take that road still receive the supply due to them; and it does not treat a child's position before the gate as the school's output. This step, extending the reading of chokepoints to the school's evaluation, is this series' own undertaking.
Two stretches, entered apart. The party that assesses the school by positions before the gate answers for its assessment; the school answers for the stretch it brought in itself. The two are entered separately. That those above assess the school by progression rates does not exempt from questioning the school's bringing preparation for the gate into every lesson; that the school brought it in does not exempt that assessment from questioning. The evaluation reform plan read in Paper 5, which requires that schools and teachers not be assessed by progression rates and that teachers and students not be rewarded with the results of the secondary-school and university entrance examinations as the standard, is one way of writing the stretch of assessment.4 By what particular indicators, and how, higher authorities carry out their assessment of schools, this paper has not yet found a suitable operating text to show.
An objection. One view runs: a uniform examination is the guardian of fairness; without a uniform score, the gate is handed over to connections and birth.5 Its force lies in this: a score is at least a number everyone has to pass, and what goes around it may go somewhere even less visible. This paper is not against tests (§2). How the gate itself is designed, in the stretch outside the school, is handed to other work on fairness and progression. The school's stretch has to bear this risk in fact: what internal evaluation has adopted, on what grounds, and how it is questioned; recommendations and streaming that can state their grounds and have a road of review (§1). To say only "not against tests" is not enough.
XX29 The gate of progression: the school answers for how much of it it brings in. Entrance examinations for the next stage often sit outside the school, and are a chokepoint; where the school's grades, comments and recommendations are adopted by the gate, and in moving up and streaming within the school, the school is inside the gate. The school answers for how it brings in and adopts the requirements of the chokepoint of progression: how much time preparation for the gate takes, and what it squeezes out; whether teaching, care, response to particular questions, and time with no preset use remain; whether students who do not take that road still receive the supply due to them; it does not treat a child's position before the gate as the school's output. The party that assesses the school by positions before the gate answers for its assessment; the school answers for the stretch it brought in itself; the two are entered separately.
§5 A Record Is a Limited Proof
It states what was verified. Report cards and comments state what the school verified, and by what materials and standards it made the evaluation; they do not vouch for what kind of person this child is. At the level of the university the Theory of the University said that a degree is a limited proof: the side that writes it states what was written in, and the side that uses it answers for its adoption.
Comments do not hang on the person. Teacher-Student Theory writes: "Saying an expectation out loud is a form of questioning. To hang an expectation on what kind of person the student is, for instance 'a good student wouldn't do this' or 'you have talent, you shouldn't be like this,' is to fix for him what something means to him, and slides in the direction of colonization". This paper extends that to the school's comments: a comment may write what he did this term and where he has not yet got; it does not hang expectations on "what kind of person you are." This step is this series' own undertaking.
Records leaving the school. Once a record leaves the school, the giving side states what it gave and on what ground, and the adopting side answers for its adoption; Paper 3 has already set this down. A rule that writes how comments are to be written, this paper has not yet found.
XX30 A record is a limited proof. Report cards and comments state what was verified and on what basis the evaluation was made, and do not vouch for what kind of person this child is; comments do not hang expectations on "what kind of person you are."
§6 Evaluation in Three Stages
This section follows the axis of stages and writes only the arrangements the school mainly handles at each stage. That a matter is written at some stage does not mean it appears only there, nor is the stage used to judge where a particular child has got to.
Kindergarten. No subject-style tests or rankings are organized for young children (Paper 1).
Primary school. No pen-and-paper examinations in the first and second grades; graded evaluation; rankings not made public.
Secondary school. The gate of progression; streaming and recommendation within the school.
Where This Paper Stops
What it has not done: it has not said that tests ought to be abolished; it has not said which grading system is best; it has not designed entrance examinations.
Notes
- For a well-known statement of this view, see Illich (1971), Deschooling Society. This paper has not gone back to the original, lists it only as a lead, and writes only its structure in the text. ↩︎
- Of the set of policies to which this notice belongs, this paper takes only the structure of the article on examinations. ↩︎
- Korea's grading of internal school records in upper secondary school into bands by fixed proportions is another way of writing grading by distribution; this is known to this paper through secondary material, and it has not gone back to the original. ↩︎
- That plan is paraphrased from a reprinted copy. ↩︎
- This paper writes only the structure of this view, and does not report any particular author's words. ↩︎
The Five Propositions of This Paper
XX26 Evaluation is power, and matters more for those who cannot walk away. The school's evaluation (marks in courses, conclusions on promotion and class assignment, comments), together with decisions that adopt evaluation, changes the roads a child can take; the school sets checks on it, and the strength of the checks follows the actual effect; not being able to walk away and not yet being able to say no both make the effect deeper. Where the school adopts automated evaluation, allocation or recommendation, it answers for its adoption. (Normative, "both make the effect deeper" being descriptive; Theory of the University DX22, which speaks of the university's evaluation, extended to children, this series' own undertaking; checks following the effect take over Political Economy E14, the chokepoint E7; the last sentence has the same shape as Paper 1, XX5, and Theory of the University DX24)
XX27 What can be measured is not all. The school may test whether its teaching has reached the children; it does not register meeting some indicator as having replaced judgment of particular work and a particular child, nor use it to exempt the evaluating position from explaining and answering. (Normative; taking over Theory of the University DX23; Chapter 1 of Education as Subject-Condition points the same way)
XX28 Not setting up one child as the measure of another. Arrangements such as public rankings, role models and model classes are read first by how they land on each child; where one child is set up as the measure of another child's whole person, the school answers for it; where it also actually alters the situation of the child used as the measure (compared in public, assigned to a class on that basis), that is entered separately. Using a piece of homework to show one way of doing something is not of this kind; that there are rankings is not in itself judged as crossing the line. (Normative; Teacher-Student Theory TS23, which speaks of a teacher's sentence, extended to the school's arrangements, this series' own undertaking)
XX29 The gate of progression: the school answers for how much of it it brings in. Entrance examinations for the next stage often sit outside the school, and are a chokepoint; where the school's grades, comments and recommendations are adopted by the gate, and in moving up and streaming within the school, the school is inside the gate. The school answers for how it brings in and adopts the requirements of the chokepoint of progression: how much time preparation for the gate takes, and what it squeezes out; whether teaching, care, response to particular questions, and time with no preset use remain; whether students who do not take that road still receive the supply due to them; it does not treat a child's position before the gate as the school's output. The party that assesses the school by positions before the gate answers for its assessment; the school answers for the stretch it brought in itself; the two are entered separately. (Normative; the first sentence and "the school is inside the gate" are descriptive; Education as Subject-Condition 4.1, and Political Economy E7 and E31, extended to the school, this series' own undertaking)
XX30 A record is a limited proof. Report cards and comments state what was verified and on what basis the evaluation was made, and do not vouch for what kind of person this child is; comments do not hang expectations on "what kind of person you are." (Normative; taking over Theory of the University DX26; Teacher-Student Theory TS20 extended to the school's comments, this series' own undertaking)
References
All SAE works below are cited by concept DOI, which always resolves to the latest version.
I · This system
- Qin, H. (2026). What a School Is Not (SAE Theory of the School, Prequel) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24382559
- Qin, H. (2026). School Teaching (SAE Theory of the School, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24382568
- Qin, H. (2026). School Care (SAE Theory of the School, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24382574
- Qin, H. (2026). Education as Subject-Condition: A Philosophy of Education [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.18867389
- Qin, H. (2026). The Place of Teacher-Student Theory (SAE Teacher-Student Theory, Paper 0) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24267256
- Qin, H. (2026). Chokepoints (SAE Political Economy, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23874140
- Qin, H. (2026). Checks (SAE Political Economy, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23874142
- Qin, H. (2026). Growth (SAE Political Economy, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23874150
- Qin, H. (2026). University Evaluation (SAE Theory of the University, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24293751
II · Rule texts
- General Office of the Ministry of Education (China). 2021. Notice on Strengthening Examination Management in Compulsory Education Schools (关于加强义务教育学校考试管理的通知), Jiaojitinghan [2021] No. 34. 2021-08-30.
- General Office of the Ministry of Education (China). 2025. Notice on Further Strengthening the Management of Routine Examinations in Primary and Secondary Schools (关于进一步加强中小学日常考试管理的通知), Jiaojiting [2025] No. 3. 2025-12-04.
- Ministry of Education, Culture, Sports, Science and Technology (Japan). Explanation of learning evaluation (on the change of grading from relative to absolute evaluation).
- Central Committee of the Communist Party of China and the State Council. 2020. Overall Plan for Deepening the Reform of Educational Evaluation in the New Era (深化新时代教育评价改革总体方案).
- Colorado. SB22-070. 2022.
III · Other literature
- Campbell, D. T. (1976). Assessing the Impact of Planned Social Change. As quoted in Theory of the University, Paper 5 (from the 2011 reprint in Journal of MultiDisciplinary Evaluation 7 (15)).
Interfaces
This series: XX26 to XX30.
Handed to Paper 7: all these arrangements cost money, public money among it.
Handed to other papers: checks on evaluation and roads of review, in Paper 8, §1, on the accountability of evaluation; the gates of admission to upper secondary school and kindergarten, Paper 7 (XX31).