Extract Research Methods from a Paper
Extract study design, sample, instruments and analysis from a paper, with verified quotations and locations for checking the original methods.
Paper
Process up to 150,000 characters. Longer papers are explicitly marked as partial coverage; inspect extracted text before processing.
Drag a file here, or
Text-layer PDF, TXT, Markdown or JATS XML; up to 25 MB. OCR is not supported. PDF page numbers differ from printed page labels.
2 papers shared across extraction presets and the summarizer. Reprocessing the same paper is free.
Paper text is sent to our backend and OpenAI. This application does not store paper text or results; provider retention terms apply. Only allowance identifiers persist. Copy or download before closing this tab.
Continue paper extraction in AnswerThis
Explore paper extraction in the AnswerThis app. Copy or download this result before leaving.
How to use Extract Research Methods from a Paper
- Paste or upload an authorized readable paper and inspect its identity and coverage.
- Review parsed text and coverage. Select extraction fields; inferred limitations are optional and off by default.
- Process the paper, then compare generated information with every exact quotation and source location.
- Copy or download the reviewed result before leaving the page.
Read design, recruitment and analysis as separate decisions
The methods preset selects methods plus sample and population. It is designed to help you locate what the paper says about study design, recruitment, measurement and analysis while keeping evidence nearby. A single design label rarely explains everything needed for interpretation. Random assignment, prospective follow-up, masking and the unit of analysis describe different aspects of a study. Do not infer one from another, or treat an experimental-sounding title as evidence that participants were randomized to a comparison.
Identify the population the authors intended to study and the sample they actually observed. Eligibility criteria, recruitment source, setting, exclusions and retention can all change who contributes data. A total enrolled number may differ from the number included in the primary analysis. Keep those stages explicit when transferring information into a review table. For clustered or repeated-measures designs, distinguish participants from clusters and observations; the extractor preserves reported descriptions rather than calculating an effective sample size.
The generated methods description is a reading aid. Its evidence quotations should point to the actual design and sample statements, but a quotation match does not establish that the summary interpreted them correctly. A proposed protocol and a completed study also need different tense and claims. An intended procedure is not evidence that it was carried out. Check the document stage and compare methods with the reported results before describing the study as having completed a particular analysis.
Check instruments, variables and statistical procedures
Measurement involves more than naming an instrument. Look for the outcome definition, units, timing, scoring, administration procedure and any adaptation for the study population. A familiar scale name does not guarantee that the same version or threshold was used across papers. For laboratory and computational work, instruments may include assays, devices, software versions, preprocessing or model-training procedures. Retain those details when they affect reproducibility. The tool does not fetch a cited measurement manual or validate an instrument from its name.
Analysis methods should be linked to the question and data structure. A paper may report descriptive comparisons, adjusted models, sensitivity analyses and exploratory work in separate places. Preserve the distinction between a primary analysis and a secondary or post hoc analysis. In an observational study, adjustment does not automatically justify causal language. In a randomized trial, the stated analysis population matters. A correctly quoted statistical-test name is only one part of the information required to understand what an estimate represents.
Look for handling of missing data, covariate selection, multiplicity, model assumptions and deviations from a planned analysis. These details can determine whether two apparently similar methods are comparable. If the tool cannot locate them in the supplied text, it should leave the relevant description uncertain rather than fill in conventional practice. Do not assume complete-case analysis, intention-to-treat analysis or a particular imputation method because it would be common in that field. Follow references to a protocol or supplement manually when needed.
Turn extracted methods into a reviewer-checked record
Compare the extracted information with your review data dictionary. Decide whether your methods column needs a short study-design label or a longer account of recruitment, intervention delivery, instruments and analysis. The selectable extractor can add outcomes, effect sizes or datasets when those fields are needed, without treating another preset as a new paper. Keep evidence and page references attached until the review team has resolved disagreements. The resulting record supports review work but is not a completed risk-of-bias assessment.
PDF text extraction may mix columns, omit figure text or separate a table header from its cells. Coverage reports how many pages yielded readable text, not whether every visual element was reconstructed correctly. The 150,000-character limit also means a long document may be processed only in part, with an explicit banner. Scanned PDFs require external OCR and encrypted documents require an authorized unlocked copy. Check the displayed text and the original tables before interpreting a missing method as absent from the study.
Check the title and methods in your supplied document before processing. Upload or paste an authorized readable paper with the complete methods and relevant supplements. Exports contain the generated information, verified evidence, locations and uncertainty states. Two papers are shared with the other presets and summarizer; failed processing costs no paper. Copy or download the reviewed result before closing this tab or opening AnswerThis, because signup does not transfer the document.
Supplied documents and processing
Upload or paste a paper you are authorized to process, then inspect its parsed text and coverage. Only supplied text is processed; this tool does not retrieve articles. Publication version and licence are not independently verified. Reports include the processing time, source description and coverage.
Paper text is sent to our backend and OpenAI. This application does not store paper text or results; only allowance identifiers persist. Provider retention terms apply. See OpenAI data controls and retention. Copy or download your reviewed result before closing the tab.
Frequently asked questions
Does the sample mean recruited or analysed participants?
Those counts can differ. Check quotations for recruitment, allocation, exclusions and the analysed sample. Keep each count attached to its stage and subgroup instead of treating every reported number as the final sample size.
Will the tool identify instrument versions?
Only versions, scoring details and procedures reported in the supplied text should be retained. Verify instrument names and versions against the methods or supplements; an unnamed outcome difference cannot establish which scale was used.
Can planned procedures be reported as completed methods?
No. A protocol or future-tense plan describes intended work. Preserve that status and distinguish it from completed recruitment, measurement and analysis. Check tense and timing against the exact quotation.
Which files and page references are supported?
Upload text-layer PDF, TXT, Markdown or JATS XML up to 25 MB. OCR is not supported. Encrypted, corrupt and scanned files need a readable replacement. PDF page numbers are separate from printed labels; JATS uses section/paragraph references with pages unavailable. Coverage is explicit, including the 150,000-character limit.
Which paper text should I supply?
Upload or paste an authorized readable paper including the sections you want processed. Check the title, body and supplements before generating. An abstract alone is not the full paper. Only supplied text is processed; the tool does not retrieve papers or bypass access restrictions.
How does the 2 papers allowance work?
This tool includes 2 papers per browser, shared across extraction presets and the summarizer. A distinct document hash counts once after successful processing. Re-running the same paper is free. Confirmed processing failures are refunded; a lost response or browser storage failure leaves allowance status uncertain. Copying or downloading an existing result is free.
Does signup save the paper or unlock more runs here?
Signup opens the AnswerThis app. It does not unlock more runs on this page or transfer your draft. Copy or download your result first. This application does not store paper text or results; only allowance identifiers persist. Paper text goes to our backend and OpenAI; provider retention terms apply.