Finding a source you can defend · 找到站得住脚的材料
| English | 中文 | Pinyin · 拼音 |
|---|---|---|
| keywords/ˈkiːwɜːdz/ | 关键词 | guān jiàn cí |
| operators/ˈɒpəreɪtəz/ | 运算符 | yùn suàn fú |
| authority/əˈθɒrɪti/ | 权威性 | quán wēi xìng |
| currency/ˈkʌrənsi/ | 时效性 | shí xiào xìng |
| bias/ˈbaɪəs/ | 偏见 | piān jiàn |
| primary source/ˈpraɪməri sɔːs/ | 一手资料 | yī shǒu zī liào |
| secondary source/ˈsekəndəri sɔːs/ | 二手资料 | èr shǒu zī liào |
Two headlines, two different claims
- One page says an electric car has no exhaust emissions; another discusses emissions from making its battery. They answer different questions.
- Before choosing a source, write the claim you need to check: total lifecycle emissions, for a stated vehicle, place and period.
两条标题，两个不同的主张
- 一页指出电动汽车没有尾气排放；另一页讨论制造电池过程中的排放。它们回答的是不同的问题。
- 在查阅资料前，先写下需要验证的主张：全生命周期排放量，并指定具体的车辆、地点和时间段。
A top-ranked search result is guaranteed to be accurate. · 排名靠前的搜索结果一定准确无误。
Ranking is not a guarantee of accuracy. Open the page and inspect the evidence. · 排名并非准确的保证。请打开网页并检查相关证据。
Turn the claim into a search
- Choose keywords 关键词 for the object, measure and comparison:
electric vehicle lifecycle emissions petrol comparison. Try everyday words too if specialist words miss useful pages. - Operators 运算符 narrow Google searches: quotes request an exact phrase,
-wordexcludes a term, andsite:example.orgrestricts a domain. Do not put a space aftersite:.
A domain filter changes where you search, not whether a claim is true. A government, university or charity page still needs a check of its method and relevance.
将主张转化为搜索词
- 为对象、测量指标和比较选择关键词：
electric vehicle lifecycle emissions petrol comparison。如果专业术语未能找到有用页面，也可以尝试日常用语。 - 运算符 用于缩小 Google 搜索结果范围：引号要求精确短语，
-word排除特定词汇，site:example.org限制域名。请勿在site:后添加空格。
域名过滤器仅改变搜索范围，并不决定主张的真伪。政府、大学或慈善机构的网页仍需对其方法和相关性进行检查。
You need evidence comparing total emissions over a vehicle lifetime. Which query best matches that claim? · 你需要关于车辆全生命周期总排放量的对比证据。哪个搜索查询最符合这一主张？
Lifecycle and comparison specify the measure and comparison. Search terms help find candidates; they do not verify them. · 生命周期与比较指明了测量指标和对比维度。搜索词有助于找到候选结果，但无法验证其准确性。
Complete the Google domain filter: ____example.org · 补全 Google 域名过滤器：____example.org
Use site: directly before the domain, without a space. · 在域名前直接使用 site:，中间不要加空格。
Open the page and test its fit
- Read beyond the search snippet. Check the publication date, location, vehicle type and what the emissions measure includes.
- Authority 权威性 concerns relevant expertise; currency 时效性 concerns whether the evidence is recent enough for this claim. Newer is not automatically better.
打开网页并检验其适用性
- 阅读内容应超出搜索摘要片段。检查发布日期、地点、车辆类型以及排放量的测量范围是否包含相关项目。
- Authority 权威性关注相关专业知识；currency 时效性关注证据是否足以支持该主张。并非越新就越好。
Match the source check to its purpose. · 将来源核查与其目的相匹配。
Each check answers a different question. Reputation alone does not establish fit. · 每一项核查都回答不同的问题。仅凭声誉不能确立适用性。
Follow the evidence, including its limits
- Bias 偏见 can arise from funding, selection or wording. Identify interests and compare independent evidence; funding alone does not prove the result is false.
- A primary source 一手资料 reports original evidence; a secondary source 二手资料 interprets it. Read the original for its methods when possible, and use a reliable review to understand several studies.
A fictional report compares two cars over 150,000 km using one electricity mix. It supports a comparison under those assumptions. It does not prove that every electric car in every country has the same result.
遵循证据，包括其局限性
- Bias 偏见可能源于资助、样本选择或措辞。识别利益相关方并对比独立证据；仅凭资金来源不能证明结果是错误的。
- Primary source 一手资料报道原始证据；secondary source 二手资料对其进行解读。尽可能阅读原始材料以了解其方法，并使用可靠的综述来理解多项研究。
一份虚构报告在150,000公里范围内使用单一电力组合比较了两款汽车。它在这些假设下支持了比较结论。但这并不能证明每个国家的所有电动汽车都有相同的结果。
A news story links to a study. You need to check the study method. What should you do first? · 一篇新闻报道链接到一项研究。你需要核查该研究方法。你首先应该做什么？
The original is the first place to check how its data was collected. A review can still help compare studies. · 原始出处是核查其数据收集方式的第一手资料。文献综述仍有助于比较研究。
A report models two cars over 150,000 km using one electricity mix. Which are justified? Select all that apply. · 一份报告使用单一电力组合对两辆汽车进行了 150,000 公里的模拟。哪些结论是合理的？请选择所有适用项。
Conditions define what a comparison supports. Neither universal claims nor automatic rejection from funding are justified. · 条件定义了比较的适用范围。既没有普遍性主张，也不存在因资金来源而自动被否定的合理性。
Make a note you can trace
- Record author or organisation, title, date, URL and the page or table used. Follow the citation style your task requires.
- Mark copied words as a quotation immediately. A paraphrase changes the wording and structure but still cites the source of the idea.
记录可追溯的笔记
- 记录作者或机构、标题、日期、URL以及所用的具体页面或表格。遵循任务要求的引用格式。
- 立即将复制的文字标记为引文。改写改变了措辞和结构，但仍需注明想法的来源。
A paraphrase of a study finding still needs a citation. · 研究发现的改写仍需注明出处。
Changing wording does not change where the idea came from. · 更改措辞不会改变观点的来源。
Build a defensible conclusion
- Compare sources on the same measure before deciding they disagree. Explain the conditions and any unresolved evidence gap.
- For your own task, keep a source log with one supported claim, its evidence location and one limitation for each source.
Search → check fit → inspect evidence → record the source → state a limited conclusion. Ranking and domain names are clues, not guarantees.
Further guidance: Google search operators.
构建有辩护力的结论
- 在决定两者相矛盾之前，先在同一指标上对比来源。说明适用条件及任何未解决的数据缺口。
- 针对你自己的任务，建立来源日志，为每个来源记录一项受支持的主张、证据位置及一个局限性。
搜索 → 核查契合度 → 检查证据 → 记录来源 → 陈述有限度的结论。排名和域名只是线索，而非保证。
更多指导：Google搜索运算符。
One source measures exhaust emissions; another measures production plus use. What resolves the apparent disagreement? · 一个来源测量尾气排放；另一个测量生产加使用阶段。什么能解决这种表面上的分歧？
These are different measures. Align what is counted before comparing the findings. · 这些是不同的衡量标准。在比较发现之前，应对统计口径进行对齐。