← Facilitator Hub教师资源中心

Assessment测评

The same instrument before Lesson 1 and after Lesson 12. Scored on reasoning, not verdicts.同一套工具,在第 1 课前和第 12 课后各做一次。评的是推理,不是结论。

How it works如何使用

Give it twice. Compare the reasoning.做两次,比较推理过程。

Each track has one assessment instrument and one scoring rubric. Administer the instrument before Lesson 1 (the pre-assessment) and the identical instrument after Lesson 12 (the post-assessment). The change between the two is the result. 每条路线有一套测评工具和一份评分表。在第 1 课前进行测评 (前测),并在第 12 课后用同一套工具再测一次(后测)。两次之间的变化即为结果。

You don't score whether the child reached the "right" verdict. You score how they reasoned — whether they named a specific suspect claim and a concrete way to check it. The rubric breaks this into dimensions and maps each one back to the lessons that build it. 你不评孩子是否得出"正确"结论,而是评他们如何推理——能否指出 某个具体的可疑说法,以及一个可行的核查方法。评分表把这些拆成若干维度,并将每个维度 对应回培养它的课程。

An honest note. This is a well-designed classroom tool, not a clinically validated psychometric test. Use it to see whether a child's reasoning grew and which lessons to reinforce — not as a research-grade score. 坦诚说明。这是一套精心设计的课堂工具,并非经临床 验证的心理测量量表。请用它来观察孩子的推理是否成长、该加强哪些课——而不是当作研究级别的分数。
The instruments测评材料

Five files五份文件

📄
Assessment overview & how-to测评总览与使用说明 How and when to administer, how to read the results, and the honesty note.如何及何时实施、如何解读结果,以及坦诚说明。
README
🟠
Primary assessment — "Robo's Answers"小学测评——"Robo 的回答" Track A. A facilitator-administered spoken interview, 5 tasks.A 线。由教师主持的口头访谈,共 5 个任务。
Track AA 线
🟠
Primary scoring rubric小学评分表 5 dimensions × 0–2 = /10, with a record sheet, score bands, and a per-dimension lesson-fix map.5 个维度 × 0–2 = 满分 10,附记录表、分数段,以及各维度对应的补强课程。
Track AA 线
🔵
Junior High assessment — "Judge This AI Answer"初中测评——"评判这个 AI 答案" Track B. A written, self-completed task: one engineered AI answer with planted issues, 6 questions.B 线。书面、学生独立完成的任务:一个植入了若干问题的 AI 答案,共 6 道题。
Track BB 线
🔵
Junior High scoring rubric初中评分表 6 dimensions × 0–3 = /18, with an answer key of planted issues, a record sheet, bands, and a course-improvement map.6 个维度 × 0–3 = 满分 18,附植入问题的答案、记录表、分数段,以及课程改进对照表。
Track BB 线
The Junior High instrument contains an engineered AI answer. Its planted issues — a fake citation, a bandwagon appeal, a false dilemma, an overclaim, a leading question — are intentional and stable; you do not need to regenerate them. They are a designed test item, not a live AI capture. 初中测评工具中包含一个特意设计的 AI 答案。其中植入的 问题——假引用、从众诉求、虚假两难、过度断言、引导性提问——是有意为之且稳定的,无需重新生成。 它是一道设计好的测试题,而非实时 AI 的截取。