Research methods guides
Step-by-step, no-fluff guides to the methods work every study runs into, written by the team building Kahubi, linked to free tools you can use right away.
How to ground AI research answers in real databases with MCP
Connect AI to PubMed, Eurostat, EUR-Lex and national archives: worked examples by field, prompts that stay grounded, and the limits to watch.
Read the guideSeriesAI guidance at universities, country by country
What universities allow in graded work, whether you may cite AI, what to disclose and which tools are safe for personal data, compared and sourced.
Compare countriesHow to write a systematic review, step by step
From a fuzzy topic to a PRISMA-compliant manuscript: the full workflow (question, search strategy, screening, extraction and write-up) with the common failure points marked.
Read the guideGuideChoosing the right statistical test
Four questions determine the test: what kind of outcome, how many groups, are measurements paired, and can you trust the normality assumption. Answer those and the choice is nearly mechanical.
Read the guideGuideHow to analyze interview data
Transcribe, code, theme, report: the qualitative analysis pipeline explained, with the discipline that separates real thematic analysis from vibes with quotes.
Read the guideGuideSample size: how many participants do you actually need?
Too few participants and real effects vanish into noise; too many and you burn months of recruitment for nothing. The answer is a calculation, not a convention. Here is how it works.
Read the guideGuideCitation styles explained: APA, MLA, Chicago, Harvard, IEEE, Vancouver
Same paper, six different reference formats. What actually differs between the major styles, which one your field expects, and how to stop formatting references by hand.
Read the guideGuideHow to write an annotated bibliography
A citation plus a paragraph that proves you read the source: what belongs in the annotation, how long it should be, and a structure you can reuse for every entry.
Read the guideGuideHow to write a literature review
A literature review is an argument about a field, not a stack of summaries. How to find the right sources, organize them into themes, and write synthesis that examiners and reviewers actually reward.
Read the guideGuideHow to write a research proposal
A proposal is a persuasion document with a technical core: convince a committee the question matters, the method can answer it, and you can do it in the time you claim. Here is the structure that works.
Read the guideGuideHow to write a methodology section
The methods section has one reader-facing job (let a skeptic judge whether your results can be trusted) and one field-facing job: let someone repeat the study. Write for both and reviewers have little left to attack.
Read the guideGuideHow to respond to reviewer comments
Major revisions are an invitation, not a rejection. The response letter is its own genre (point-by-point, gracious, and precise), and doing it well is one of the most learnable skills in academic publishing.
Read the guideGuideHow to use AI for research, responsibly
AI is now part of the research workflow whether institutions like it or not. The productive question is no longer whether to use it, but which tasks to hand over, which to keep, and how to stay on the right side of your journal, your university, and your own methods section.
Read the guideGuideHow to write an abstract
The abstract is the most-read 250 words of your paper, and often the only ones. It is also a genre with rules: five moves, in order, no suspense. Here is how to write one that gets the paper read, cited, and accepted.
Read the guideGuideHow to write a discussion section
The discussion is where papers are won and lost in review: results only become a contribution when you say what they mean, square them with the field, and admit what they cannot show. A structure that keeps you honest and persuasive.
Read the guideGuideHow to transcribe research interviews in Swedish
Which tools handle Swedish, where they process your recordings, how accurate they are, and what GDPR asks of you. Facts checked on each provider's own pages on 7 October 2026.
Read the guideGDPR guideHow to transcribe research interviews with AI, GDPR compliant
AI transcription saves days of work per study, but an interview recording is personal data about someone who trusted you. This guide covers what GDPR asks of you, how to choose a service, and a workflow from consent form to coded transcript.
Read the guideGDPR guideHow to transcribe patient interviews with AI under GDPR
Patient interviews are among the most valuable and most sensitive data in research. Health information is a special category under GDPR, so the rules are stricter. Here is how to get accurate transcripts without putting patients or your approval at risk.
Read the guideGDPR guideHow to transcribe focus groups with AI, GDPR compliant
Focus groups are the hardest recordings to transcribe: six to ten voices, overlapping talk and side comments. They also raise their own privacy questions, because participants hear each other. Here is how to record them well, transcribe them in the EU and turn them into analysable data.
Read the guideGDPR guideHow to transcribe research meetings with AI under GDPR
Lab meetings, supervision sessions, advisory boards and stakeholder workshops produce decisions and ideas that are easy to lose. AI transcription can capture them, but it is still personal data about the people in the room. Here is how to do it fairly and usefully.
Read the guideResearch integrityHow to disclose AI use in a research paper, with examples
Using AI in research is allowed almost everywhere. What journals and funders ask for is honesty about how you used it. Here is what the main policies say, what you do not need to disclose, and example statements you can adapt.
Read the guideResearch integrityHow to check AI-generated references for fake citations
AI can save hours of searching, but general chatbots still invent references that look real. Here is what the studies show, why it happens, and a fast routine to check every citation before it reaches your reference list.
Read the guideResearch integrityWhy AI detection tools don’t work as evidence, and what to do instead
AI detectors promise to tell human writing from machine writing. The newest ones have improved, but the research is clear that a detector score cannot show how a real text was written. Here is the evidence, what librarians, integrity researchers and universities now recommend, and what the European Commission’s guidelines ask for instead.
Read the guideGuideAI tools for literature reviews: what they can and cannot do
AI search tools can find relevant papers in minutes, and some surface studies a keyword search misses. Published evaluations also show where they fall short. Here is an honest map of the tools, what librarians found when they tested them, and how to use them well.
Read the guideGuideHow to use AI in a systematic review responsibly
The major evidence synthesis organisations now accept AI in systematic reviews, on clear conditions. Here is what their 2025 joint position statement says, what the evidence shows for screening, extraction and searching, and how to report it.
Read the guideQualitative researchAI-assisted thematic analysis: is it valid, and how to do it well
AI can suggest codes in minutes, and studies show it often finds the same broad themes as human analysts. It also misses subtext and context, and leading qualitative methodologists have rejected it for reflexive work. Here is the evidence on both sides and a practical way to decide.
Read the guideGDPR guideCan you upload research data to ChatGPT? A GDPR guide for researchers
AI assistants are excellent for analysis and writing, but research data often includes personal data about participants. Here is what GDPR requires, how consumer and business AI accounts differ, what universities are telling their researchers, and how to use AI safely.
Read the guideGDPR guideHow to anonymise interview transcripts
Removing names is not enough. People can be recognised from their job, their town or a single unusual story. This guide explains the difference between pseudonymisation and anonymisation, the techniques data archives recommend, and what AI redaction can and cannot do.
Read the guidePut the methods to work
Every guide maps to a Kahubi flow that runs the workflow end to end. Free plan included.