Add research workflow guide

README.md

@@ -1,3 +1,80 @@
-# supplementary-materials-guide
+# Managing Supplementary Materials So Nothing Important Gets Missed
 
-Practical academic research workflow guide by Yila AI
+Supplementary materials have grown from a few extra tables into, in many papers, the larger part of the work. Full protocols, sensitivity analyses, item-level questionnaires, additional outcomes and code often live in supplementary files that many readers never open. That creates a real risk: the details that decide whether a finding is robust, or whether a method can be reused, are hidden in plain sight. This guide explains how to read and manage supplementary materials efficiently, and how an evidence-traceable research workbench such as [Yila](https://yila.ai) helps you find what matters in them.
+
+## Why supplements matter more than they used to
+
+Word limits in the main text have pushed essential information into supplements. It is now common for the main paper to report a primary analysis while the supplement contains the sensitivity analyses that test it, the subgroup results that qualify it, and the protocol deviations that complicate it. Peer reviewers do not always scrutinise supplements as closely as the main text, so errors and inconsistencies there can survive publication.
+
+If you cite, build on or appraise a paper without checking its supplement, you are working with part of the evidence.
+
+## What to look for in a supplement
+
+Not every supplement needs to be read cover to cover. Focus on the items most likely to change your interpretation:
+
+1. **The protocol and statistical analysis plan.** Compare the planned outcomes and analyses with those reported.
+2. **Sensitivity analyses.** Do the main results hold under alternative assumptions, such as different handling of missing data or different model specifications?
+3. **Subgroup and secondary results.** Are effects consistent across groups, or concentrated in one?
+4. **Participant flow details.** How many people dropped out, when and why?
+5. **Measurement details.** Exact questionnaire items, scoring rules and validation information.
+6. **Adverse events or harms.** Often reported in full only in the supplement.
+
+## Keep supplements with the main paper
+
+A practical problem with supplements is that they get separated from the paper. You download the main PDF, intend to fetch the supplement later, and never do. Make it a habit to download supplementary files at the same time as the paper and store them together, with consistent file names that link them.
+
+A research workbench that holds the main text and supplements in one place makes this easier and pays off when you start asking questions across both.
+
+## Ask targeted questions across all files
+
+The main barrier to using supplements is time. A long supplement may run to dozens of pages of tables with minimal explanation. Searching it manually for a specific detail is slow.
+
+An evidence-traceable assistant changes this. With the paper and its supplements uploaded together, you can ask Yila a direct question such as "Were the results robust to multiple imputation of missing outcome data? Show where this is reported." The answer comes back with a link to the specific table or paragraph, whether it sits in the main text or on page 37 of the supplement. You open the location, read it in context, and record your conclusion.
+
+This approach also catches discrepancies. If the main text says a sensitivity analysis "confirmed the primary result" and the supplementary table shows a confidence interval crossing the null, you will see both passages side by side.
+
+## Record supplementary evidence precisely
+
+When you extract information from a supplement, record its location precisely: file name, table or figure number, and page. "Supplement 2, eTable 4, p. 12" is far more useful than "supplement". Precise locations make your notes checkable and save time when you revisit the paper.
+
+In evidence tables, consider a separate column for supplementary sources, so you and your co-authors can see at a glance which findings depend on supplementary material.
+
+## Watch for versioning problems
+
+Supplements are sometimes updated after publication, and occasionally the online version differs from the one reviewed. If a detail matters a great deal, note the date you accessed the supplement and, if possible, keep a copy. When a correction notice mentions the supplement, check which values changed.
+
+## An example of what a supplement can change
+
+Consider a trial whose main text reports that an exercise programme reduced fatigue scores at twelve weeks. The supplement shows that the effect was driven by participants who attended more than three quarters of sessions, that a per-protocol analysis was presented alongside the intention-to-treat analysis, and that fatigue was measured with a modified version of a validated scale. None of these details makes the trial worthless, but each changes how you would describe it in a review or a grant. Without opening the supplement, you would have cited a cleaner result than the evidence supports.
+
+## Supplements in your own papers
+
+The same principles apply when you write. Readers will struggle with your supplement unless you make it navigable:
+
+- Start with a table of contents.
+- Number tables and figures clearly and refer to them by number in the main text.
+- Explain each table briefly: what it shows and how it relates to the main results.
+- Include the protocol and analysis plan, with deviations marked.
+- Provide code and data, or explain clearly how to access them.
+
+A well-organised supplement signals care, makes peer review easier and increases the chance that others can build on your work.
+
+## Supplements in peer review
+
+If you review papers, make a habit of checking the supplement for the items listed above, especially the protocol, sensitivity analyses and participant flow. Ask authors to move essential information into the main text if it is critical to interpreting the results. Located evidence helps here too: a review comment that points to "eTable 6, row 3" is more actionable than a general concern.
+
+## A short routine
+
+For each important paper you read:
+
+1. Download the supplement with the paper.
+2. Skim the supplement's table of contents.
+3. Ask targeted questions about the protocol, sensitivity analyses, subgroups and flow.
+4. Confirm each answer at its location.
+5. Record findings with precise locations.
+
+This routine adds perhaps fifteen to thirty minutes per paper and often changes how much weight you give to its conclusions.
+
+## Summary
+
+Supplementary materials now hold much of the evidence that determines whether a finding is robust or reusable. Keep them with the main paper, focus on protocols, sensitivity analyses, subgroups, flow, measurement and harms, record locations precisely, and make your own supplements navigable. An evidence-traceable research workbench like [Yila](https://yila.ai) makes supplements practical to use by locating answers across the main text and every supplementary file, so nothing important stays hidden.