Перейти к содержанию

С чего начать · Начальный

Template 3: Search log and evidence matrix

Use a separate row or record for each search and each study. Keep exact quotations visibly separate from your own interpretation. For systematic reviews, adapt this to the full review method and retain complete exports.

3 мин чтения · 26 окт. 2026 г.

Эта страница ещё не переведена, поэтому вы читаете английскую версию.

Главное

  • Adapt the prompts to the actual study.
  • Complete every statement from verified records.
  • Follow the target journal and institutional requirements.
Содержание

Use a separate row or record for each search and each study. Keep exact quotations visibly separate from your own interpretation. For systematic reviews, adapt this to the full review method and retain complete exports.

Use a separate row or record for each search and each study. Keep exact quotations visibly separate from your own interpretation. For systematic reviews, adapt this to the full review method and retain complete exports.

Search record

Question and concepts: [Question; concept groups and synonyms]

Database or source: [Name and platform]

Search date: [Date]

Exact search: [Copy the actual query, including field tags]

Limits and rationale: [Language, dates, design or other restrictions]

Records retrieved: [Count]

Export location: [File and format]

Additional discovery: [Citation searching, expert suggestions or other route]

Study record

Verified citation and identifier: [Authors, title, year, journal and DOI or stable location]

Version and status: [Published article, preprint, correction or other status]

Question: [Exact aim]

Design and setting: [Method and context]

Independent sample and recruitment: [Number, unit and selection]

Exposure or intervention and comparator: [Definitions]

Outcome and timing: [Measurement and units]

Main result: [Estimate, uncertainty and relevant denominator]

Analysis: [Model, adjustment, multiplicity and dependence handling]

Limitations: [Specific issues affecting interpretation]

Claim supported: [The sentence this source can justify]

Claim not supported: [An overstatement to avoid]

Exact quotation, if needed: [Quotation marks plus page or section]

My synthesis note: [Your original interpretation]

Related reports of the same study: [Links to prevent duplicate counting]

Cross study comparison

Theme or question Supporting evidence Contrary or uncertain evidence Method differences Defensible synthesis
[Theme] [Study identifiers] [Study identifiers] [Specific differences] [Bounded conclusion]

Before citing a study, open the source and check that the actual passage supports your sentence. Bibliographic software does not perform this evidential check for you.

Tutorial: build a claim ledger alongside the search log

Read Article 2 and Article 12. Give every important claim its own row, even when several claims come from the same paper. Record the exact outcome, population and comparison supporting that claim. Include the page, figure, table or section used to check it. Keep your interpretation separate from the source's reported observation.

Hypothetical comparison: One education paper measures immediate quiz performance after voluntary tutorial attendance; another measures end of term performance after random allocation to a tutoring offer. These papers do not estimate the same quantity. The matrix should preserve the different assignment processes, outcome timings and comparison groups. A synthesis could then discuss short term performance and longer term attainment separately instead of averaging unlike percentages.

Add these fields when useful: Study identifier; linked reports; claim supported; exact source location; primary or secondary source; design limitation; correction or retraction status; and the sentence in your draft that will use the evidence. In a review, track multiple reports from the same study so they do not become duplicate evidence.

Completion check: Choose three sentences from your draft at random. Can you retrieve the supporting passage, explain why the source is appropriate and identify one boundary on the claim? If not, strengthen the source record before adding more citations.