本课为原创大学书面英语能力课程，不对应统一的人教版大学教材或固定大学年级；不含听力。练习目标只反映本站材料掌握情况，不保证实际考试分数。
学习目标
在抽象说明文中联合词性、逻辑转折和词汇复现完成词库填空。
解释指标、目标和行为之间的关系，避免局部通顺但全篇矛盾的答案。
在本课主题中理解新增词块的语境用法，并把其中相关词项迁移到新的完整句子。
核心讲解
本课仍为15词10空，词形不变，每词最多用一次。提高点在逻辑而非生僻：指标可被改善，但真正目标未必实现。先标出However、Instead、rather than等转折；再检查“衡量什么”与“希望做到什么”是否相同。
词库：A visible B rarely C incentives D delay E accurate F distinguish G expand H interpreted I narrow J responsibility K resources L rapidly M outcomes N unstable O overlook
原创选词填空：When a number becomes a target
A service office wanted to answer requests more quickly. It began reporting the number of messages each employee closed in a day. The number was easy to collect and made individual performance highly (1) ___. Managers could compare days without reading every message. At first, the measure appeared to provide an (2) ___ picture of how much work had been completed.
However, some requests required more time than others. An employee who helped one person solve a complicated problem might close fewer messages than a colleague who dealt only with simple questions. If the office failed to (3) ___ between these tasks, the comparison could be misleading. The target also created (4) ___ to close messages quickly, even when the underlying difficulty had not been resolved.
The problem was not that measurement was useless. Rather, a (5) ___ measure was being asked to represent a broad goal. Counting closed messages did not show whether users understood the answer or could complete the next step. Managers therefore added a small follow-up survey about actual (6) ___. They also asked employees to explain cases in which extra time had been necessary.
These changes required (7) ___, including staff time and a modest budget. The office could not examine every case in detail, so it selected a sample. Results were (8) ___ alongside written comments rather than treated as a complete judgement of each employee. The managers recognised that a single number could (9) ___ important differences. A useful measurement system should guide questions, they concluded, not replace the human (10) ___ for deciding what good service means.
例题解析
独立示例：The indicator is easy to count; nevertheless, it may miss the main goal. nevertheless表示转折。若前半“易计数”后半“可能漏目标”，不能用therefore把计数方便当成必然漏目标的原因。词库题也要检查这个逻辑层。
本课词汇与搭配
下列为本课实际讲解的词项；词形变化及同一词的不同义项不等于新增独立词汇量。
| 词项 | 本课义项及用法 |
|---|---|
| indicator | n. 指标；a performance indicator |
| incentive | n. 激励；create incentives |
| underlying | adj. 潜在／根本的；an underlying difficulty |
| resolve | v. 解决；resolve a problem |
| represent | v. 代表；represent a broad goal |
| outcome | n. 结果；actual outcomes |
| resource | n. 资源；allocate resources |
| interpret | v. 解读；interpret results |
| distinguish | v. 区分；distinguish between tasks |
| nevertheless | adv. 然而；注意句间标点 |
词汇工坊
本表补足本课话题的读写词块，含中学承接和大学应用，不将每项都计为大学新词。先遮住中文解释推测例句，再用自己的句子替换人物、时间或对象。理解义项、掌握搭配、能够迁移是三项不同任务；下方两道迁移练习检验部分关键词块，不代表全部词项已自动掌握。
| 词项 | 义项与搭配 | 原创语境例句 |
|---|---|---|
| metric | 量化指标；a performance metric | A performance metric should relate to the intended goal. |
| proxy | 替代指标；a proxy for quality | The number of closed messages is only a proxy for service quality. |
| benchmark | 比较基准；set a benchmark | Set a benchmark before interpreting the change. |
| target-driven | 目标驱动的；target-driven behaviour | Target-driven behaviour can improve a count without improving a service. |
| distort | 扭曲；distort incentives | A narrow target may distort incentives. |
| optimise | 优化；optimise a process | Optimising one process may create difficulties elsewhere. |
| aggregate | 汇总；aggregate results | Aggregate results can hide differences between groups. |
| disaggregate | 分组拆看；disaggregate the data | Disaggregate the data when different groups face different barriers. |
| qualitative | 定性的；qualitative comments | Qualitative comments explain difficulties that the count misses. |
| quantitative | 定量的；quantitative measures | Quantitative measures are useful when their meaning is clear. |
| measurement error | 测量误差；check measurement error | Check measurement error before treating a small difference as important. |
| unintended consequence | 非预期后果；an unintended consequence | An unintended consequence of the target was hurried replies. |
| judgement | 判断；exercise judgement | Staff still need to exercise judgement in complex cases. |
| performance | 表现；evaluate performance | Evaluate performance against more than the easiest number to collect. |
练习
先独立完成，再核对答案；开放题须按要点人工评分，不以是否逐字等同范文评分。
1
填第1空，并说明系动词/使役结构对词性的限制。
2
填第2空，核对冠词和语义。
3
填第3空，解释后面的between。
4
填第4空，解释为何目标会改变员工行为。
5
填第5空，找出与broad构成对照的词义。
6
填第6空，解释新调查关注的不是仅仅发了多少消息。
7
填第7空，指出including后例子所属上位概念。
8
填第8空，说明语态。
9
填第9空，结合最后一句说明动词意义。
10
填第10空，说明for deciding的逻辑关系。
11
词汇迁移：译：一个替代指标可能扭曲激励，因此评估表现时也需要定性意见。
12
词汇迁移：译：汇总结果可能隐藏差异；分组查看数据有助于发现非预期后果。
参考答案与解析 · 完成练习后展开
参考答案与解析
1
A visible。made individual performance highly visible，使表现可见；highly修饰形容词。数据上报让别人可比较，不是“表现不稳定”。
2
E accurate。an accurate picture为“准确描述”，appeared to说明只是最初表象；unstable虽形容词但不是此处对计数可信度的评价。
3
F distinguish。distinguish between A and B区分复杂任务和简单任务；to后动词原形。
4
C incentives。created incentives to…创造做某事的激励；以关闭量为目标可能鼓励快关闭。
5
I narrow。narrow measure对broad goal，指标范围窄但目标广。
6
M outcomes。actual outcomes关注用户是否理解、能否执行下一步，是实际结果而非关闭数量。
7
K resources。人力时间和预算都是资源；复数resources与多种投入对应。
8
H interpreted。were interpreted是被动，“结果结合书面意见被解读”；不能把结果当主动解释者。
9
O overlook。单个数字可能忽视重要差异；could后原形。expand或delay都不符合“漏掉差异”的论述。
10
J responsibility。human responsibility for deciding，决定好服务标准的人类责任。完整键A E F C I M K H O J；不用B D G L N。
11
A proxy measure may distort incentives, so evaluating performance also requires qualitative comments. evaluating短语单数主语，requires正确。6分：信息3，搭配2，一致1。
12
Aggregate results may hide differences; disaggregating the data can help reveal unintended consequences. aggregate此处形容词，disaggregating为主语。6分：信息3，词块2，句法1。