Research & Literature prompts

Help me research this topic systematically.

TOPIC: [WHAT YOU WANT TO UNDERSTAND]
WHY: [WHAT DECISION OR OUTPUT THIS FEEDS]
WHAT I ALREADY KNOW: [YOUR CURRENT UNDERSTANDING]
HOW DEEP: [a working understanding / enough to write about it / enough to make a large decision]
TIME AVAILABLE: [HOURS OR DAYS]

Produce:

1. THE QUESTION, SHARPENED - restate what I am actually trying to find out. Broad topics produce broad reading; a specific question produces an answer. If my topic is too broad for my stated time, narrow it and say so.

2. THE SUB-QUESTIONS - break it into the specific things you would need to know. Ordered, because some answers determine which others matter.

3. WHAT KIND OF ANSWER EXISTS - be honest about this before I start reading. Is this a settled factual matter, an area of genuine expert disagreement, a matter of definition, or something where good data does not exist? This determines whether I am looking for the answer or for the range of positions.

4. SOURCE TYPES IN ORDER - for this specific topic, where the good information is and in what order to approach it: primary sources and original data, academic or technical literature, industry or regulatory publications, practitioner accounts, and journalism. Note which source types are unreliable for this topic and why.

5. WHO WOULD KNOW - the kinds of people or organisations with direct knowledge, and what each would be able to tell me that documents cannot.

6. THE SEARCH STRATEGY - the specific terms to use, including the technical vocabulary I may not know. Name the jargon, because searching in the wrong vocabulary is the most common reason research stalls.

7. THE CONFLICT MAP - where I should expect to find disagreement, and what usually underlies it: different definitions, different data, different time periods, or different interests.

8. WHAT TO BE SCEPTICAL OF - the claims in this area that get repeated without evidence, and any figure widely cited that should be traced to its source.

9. WHEN TO STOP - the signal that further reading is not adding anything, so I do not research indefinitely.

10. THE NOTE STRUCTURE - how to record findings so they are usable later: claim, source, date, confidence, and what it contradicts.

Help me assess this source.

THE SOURCE:
"""
[PASTE THE CONTENT OR THE KEY CLAIMS]
"""

WHAT IT IS: [study / article / report / blog post / industry publication]
WHO PUBLISHED IT: [ORGANISATION]
WHEN: [DATE]
WHAT I WANT TO USE IT FOR: [YOUR PURPOSE]
THE SPECIFIC CLAIM I CARE ABOUT: [THE CLAIM]

Produce:

1. WHO PRODUCED IT AND WHY - the publisher, their expertise, their funding, and their interest in the conclusion. A study funded by a party with a stake is not automatically wrong, but it changes how much independent corroboration you should want.

2. THE CLAIM I CARE ABOUT, TRACED - is this the source's own finding, or is it citing someone else? Follow the chain as far as the text allows. A remarkable number of widely-cited figures trace back to a single unsourced estimate, and the citation chain is where that becomes visible.

3. THE METHOD - if it is research: what was measured, how, on whom, how many, over what period. Assess whether the method supports the claim being made. Frequently the study shows something narrower than the headline says.

4. THE GAP BETWEEN FINDING AND HEADLINE - compare what the data supports to what the summary, press release or article claims. This gap is where most misinformation originates, and it is often introduced downstream of the researchers.

5. RED FLAGS - no method described, no sample size, a percentage with no base, 'studies show' with no citation, a single study presented as settled, a correlation described causally, a survey of the publisher's own customers presented as market research, or a figure that has been repeated for years without a traceable origin.

6. RECENCY - is it still current? For fast-moving topics, what may have changed since.

7. WHAT IT ACTUALLY SUPPORTS - the narrowest claim this source genuinely backs. Often much weaker than what it is being used for.

8. WHAT WOULD CORROBORATE OR CONTRADICT IT - what else to look for before relying on it.

9. THE VERDICT - use it as evidence / use it with a caveat / use it as an illustration only / do not use it. With reasoning.

I cannot verify the source's underlying data and neither can you. Say clearly where your assessment is based only on the text I provided.

Synthesise this research.

SOURCES:
"""
[PASTE EACH, SEPARATED BY --- WITH A LABEL AND DATE]
"""

THE QUESTION: [WHAT YOU ARE TRYING TO ANSWER]
WHO THIS IS FOR: [AUDIENCE]
WHAT I WILL DO WITH IT: [PURPOSE]

Produce:

1. THE ANSWER, IF THERE IS ONE - what the weight of evidence supports, stated directly, with the confidence the evidence justifies. If the sources do not converge, say that rather than manufacturing a consensus.

2. WHERE THEY AGREE - substantively. Note where sources agree on a conclusion but for different reasons, and where apparent agreement is actually one source citing another rather than independent confirmation. Two sources drawing on the same original are one data point.

3. WHERE THEY DISAGREE - and crucially, why:
   - Different definitions of the key term
   - Different populations, periods or scope
   - Different methods
   - Genuine disagreement about interpretation of the same evidence
   - Different interests
   Most apparent contradictions in research are definitional rather than substantive. Identify which these are.

4. THE QUALITY WEIGHTING - the sources differ in evidential strength. Say which carry more weight and why. Do not treat a rigorous study and a vendor blog post as equal voices.

5. THE SOURCE MAP - a table: Source | Date | Type | Its claim on my question | Strength | Independent or derivative

6. WHAT NONE OF THEM ADDRESS - the gap in this body of evidence.

7. HOW SETTLED IS THIS - is this an area with a clear answer, an area of live disagreement, or an area where the evidence is too thin to support any confident claim? Say which.

8. WHAT WOULD CHANGE THE PICTURE - the evidence that would resolve the disagreement.

9. THE HONEST SUMMARY for my audience - what to say, including the uncertainty. Do not flatten hedged findings into confident claims; that is how research becomes misinformation in one step.

Quote directly for every disagreement.

Help me find the gap in what is known about this.

TOPIC: [THE AREA]
WHAT I HAVE READ:
"""
[PASTE SUMMARIES OR KEY POINTS OF WHAT YOU HAVE FOUND]
"""
WHY I WANT THE GAP: [research project / article angle / product opportunity / decision]
MY OWN ACCESS OR ADVANTAGE: [data, experience, or access you have that others do not]

Produce:

1. WHAT IS WELL ESTABLISHED - the settled parts, so I do not waste effort re-establishing them.

2. THE TYPES OF GAP present here - be specific about which kind, because each implies a different response:
   - EVIDENCE GAP: a question people assert answers to without data
   - POPULATION GAP: established for one group, assumed for others
   - CONTEXT GAP: established in one setting, untested elsewhere
   - METHOD GAP: studied in a way that cannot answer the real question
   - RECENCY GAP: established under conditions that have since changed
   - DEFINITION GAP: people are arguing past each other because the term means different things
   - SYNTHESIS GAP: the pieces exist but nobody has assembled them
   - PRACTITIONER GAP: people who do this daily know something that is not written down

3. THE SPECIFIC GAPS - for each, what is missing, why it has not been filled (usually it is hard, unfunded, or unglamorous), and what would fill it.

4. THE GAPS I COULD ACTUALLY FILL - given the access and advantage I described. Most gaps require resources I do not have; the useful list is the short one I could address.

5. THE SYNTHESIS OPPORTUNITY - often the most accessible gap. Where the individual pieces exist but nobody has connected them, that is genuinely new work and requires no new data.

6. THE 'WHY HAS NOBODY DONE THIS' CHECK - for each gap, the honest possibilities: it has been done and I have not found it, it is much harder than it appears, it was done and the result was uninteresting, or nobody has needed it. Address this before investing.

7. WHAT I MIGHT BE MISSING - the search terms, source types or communities that would reveal existing work I have not found. Absence of evidence in my reading is not evidence of absence.

8. THE MOST PROMISING ONE, with reasoning.

Help me structure notes from this reading.

WHAT I READ:
"""
[PASTE THE SOURCE OR YOUR ROUGH NOTES]
"""

SOURCE DETAILS: [AUTHOR, TITLE, PUBLICATION, DATE, URL]
WHAT I AM RESEARCHING: [YOUR PROJECT]
HOW I WILL USE THIS: [write about it / decide something / build on it]

Produce structured notes with:

1. THE SOURCE RECORD - full citation details, captured now. Retrieving a citation six months later costs far more than recording it took.

2. THE ONE-LINE SUMMARY - what this source says, in my own words. Not the abstract.

3. THE CLAIMS - each substantive claim, with:
   - The claim in plain words
   - Whether it is this source's own finding or something it cites, and from where
   - The evidence given
   - The page or section, so I can return to it
   - My confidence in it

4. THE QUOTES - the passages worth quoting verbatim, marked clearly as quotes with their location. Keep the distinction between quotation and paraphrase absolute; a paraphrase that drifts into your notes as a quotation is how accidental plagiarism happens.

5. MY OWN THINKING - kept visibly separate from what the source says. Use a marker. In six months you will not remember which thoughts were yours and which were the author's, and that confusion is the main failure mode of research notes.

6. WHAT THIS CONTRADICTS OR CONFIRMS in what I have already read. Record the relationship now while you remember it.

7. THE QUESTIONS IT RAISES - what to look into next.

8. RELEVANCE TO MY PROJECT - which part of my work this bears on. Notes filed with no connection to a purpose never get used.

9. WHAT I WOULD NEED TO CHECK before relying on this.

Rules:
- Never record a claim without its source, even when it seems obvious at the time
- Mark inference and opinion distinctly from the source's statements
- Write in your own words except for deliberate quotes; summarising in your own words is where understanding happens

Help me design a study.

THE QUESTION: [WHAT YOU WANT TO FIND OUT]
WHY IT MATTERS: [THE DECISION]
WHO OR WHAT I CAN STUDY: [YOUR ACCESS - customers, users, documents, your own operations]
RESOURCES: [TIME, MONEY, PEOPLE]
WHAT I THINK THE ANSWER IS: [YOUR HYPOTHESIS]

Produce:

1. THE ANSWERABLE VERSION of my question. Most research questions as first stated are too broad to answer. Narrow it to something a study could settle, and say what you narrowed away.

2. WHAT WOULD COUNT AS EVIDENCE - before designing anything: what result would convince me the answer is yes, and what would convince me it is no? If no realistic result would change my mind, the study is not worth running.

3. THE DESIGN - given my access and resources:
   - What you measure, precisely defined
   - Who or what is included, and who is excluded
   - How many you need. Show rough reasoning; an underpowered study wastes the effort entirely.
   - The comparison. Evidence requires a comparison: before and after, treated and untreated, or against a benchmark. A study with no comparison group cannot support a causal claim.
   - The period

4. THE CONFOUNDERS - what else could produce the result you observe. For each, whether the design controls for it, and if not, how to at least measure it.

5. THE SELECTION PROBLEM - how the people or cases in the study differ from the population you care about. This is the limitation of almost all practical research and it should be stated, not hidden.

6. WHAT COULD GO WRONG - dropout, non-response, the measurement changing behaviour, seasonal effects, and contamination between groups.

7. THE ANALYSIS PLAN - decided now, before collecting anything. What you will calculate and how you will interpret each possible result. Deciding this afterwards is how studies find whatever the researcher hoped for.

8. THE HONEST SCOPE - what this study will and will not be able to conclude. State it plainly so the findings are not over-extended later.

9. THE CHEAPER VERSION - is there a smaller thing that would answer most of the question? Usually there is, and it is usually the right place to start.

10. ETHICS AND CONSENT - if this involves people, what they should be told and what they must agree to.

Write up my research.

WHAT I INVESTIGATED: [THE QUESTION]
HOW I DID IT: [YOUR METHOD]
WHAT I FOUND: [YOUR RESULTS AND DATA]
WHAT I EXPECTED: [YOUR PRIOR]
LIMITATIONS I AM AWARE OF: [WHAT YOU KNOW IS WEAK]
AUDIENCE: [WHO READS THIS AND WHAT THEY WILL DO]
FORMAT AND LENGTH: [WHERE IT GOES]

Produce:

1. THE FINDING - first, in one sentence with the key number. Not the background, not the method.

2. WHY IT MATTERS - what changes as a result, in the audience's terms.

3. WHAT I DID - enough for someone to judge the quality and, in principle, repeat it. Include: what was measured, on whom or what, how many, over what period, and what was excluded. Be specific; a vague method section means the findings cannot be assessed.

4. THE RESULTS - with the actual numbers, including the ones that do not support the conclusion. Reporting only supporting results is the most common form of unintentional dishonesty in research writing.

5. WHAT I EXPECTED VERSUS WHAT I FOUND - state your prior and whether it held. Where the data contradicted you, say so plainly; it is the most credible thing in any write-up.

6. ALTERNATIVE EXPLANATIONS - the other things that could account for this result, and which you can and cannot rule out.

7. LIMITATIONS - mine plus any I missed. Sample, selection, period, measurement, and generalisability. State them fully; a reader who finds a limitation you did not mention discounts everything else.

8. WHAT THIS DOES NOT SHOW - explicit, because it will be over-extended otherwise. Name the conclusions people will wrongly draw.

9. THE CONFIDENCE - how much weight this should carry. Be calibrated, not modest for the sake of it and not confident for effect.

10. WHAT WOULD STRENGTHEN OR OVERTURN IT.

11. THE DATA - what you are publishing alongside so others can check. Research that cannot be checked is an assertion.

Rules:
- Hedge where the evidence is weak and state plainly where it is strong. Uniform hedging is as misleading as uniform confidence.
- No conclusion beyond what the method supports
- Every number traceable to the method described