Many research questions arrive as broad statements that mix several separate assertions. Treating the entire statement as one unit often leads to unfocused searches and scattered notes. Segmenting the question into its component claims allows each assertion to receive its own evidence trail and a defined completion point.
The process begins with a single reading that lists every factual statement embedded in the original question. Each listed item then receives its own short label, a required evidence type and a source category. This division turns an open-ended enquiry into a set of bounded tasks that can be completed one at a time.
Extracting individual claims from the initial question
Read the question once without taking notes. On a second pass, underline or highlight every phrase that makes a verifiable statement about a person, date, event, regulation or statistic. Convert each highlighted phrase into a short declarative sentence that stands alone. For example, a question about changes in data-protection rules may contain separate claims about effective dates, responsible authorities and scope of application.
Record these sentences in a simple table with three columns: claim text, evidence type needed and status. Keep the wording close to the original so later checks can confirm whether the claim has been addressed. Avoid combining two statements into one row even when they appear in the same sentence.
Assigning evidence requirements to each claim
Decide for every claim what form of primary or official record would settle it. A date claim may need the text of a statutory instrument; a scope claim may need the current guidance page of the responsible regulator. Write the chosen evidence type directly into the table row. This step prevents later drift toward secondary summaries that restate rather than confirm the original point.
Where an official source exists, note its exact location rather than a general domain. For UK material this usually means legislation.gov.uk for statutes or the relevant .gov.uk department page for guidance. Record the precise URL and the date the page was last checked.
Ordering tasks by dependence
Some claims rest on others. A claim about the current wording of a rule cannot be checked until the claim about which rule applies has been settled. Reorder the table so that independent claims appear first. Dependent claims move lower in the list and carry a note of the claim they rely upon. This sequence reduces repeated visits to the same sources.
Review the ordered list for any claim that still contains two separate assertions. Split these before proceeding. The goal is a list in which each row can be marked complete with a single piece of evidence.
Setting completion criteria
For each row define what will count as sufficient evidence. A completion note might state “full text of regulation paragraph X viewed on legislation.gov.uk” or “current guidance section Y dated after the claim period”. Write the criterion in the same row so that later review can judge whether the task is finished without reopening the original question.
Include a column for the date the evidence was captured and the exact file name or bookmark used to store it. Consistent capture details allow another person to repeat the verification if required.
Common pitfalls in scope definition
One frequent error is to retain the original broad phrasing as the task description. This produces vague searches and makes it difficult to know when enough material has been gathered. Another is to accept any matching webpage without checking whether the publisher is the body legally responsible for the information. A third is to move on to the next claim before recording the precise location of the evidence for the current one.
These problems are avoided by keeping the table as the single working document. Any search that does not directly address a listed row is set aside until the current row is closed.
Review points during the work
After completing half the rows, reread the original question to confirm that every embedded claim remains on the list. Add any that were missed in the first extraction. At the same time, remove any row whose evidence has shown the original statement to be outside the intended scope.
At the end of the exercise the table itself becomes the record of how the broad question was reduced to manageable parts. The completed rows show which claims rest on official sources and which remain open because no suitable record was located.
Limitations of the method
The approach works best when the question contains discrete factual assertions. Questions that ask for interpretation, prediction or comparative judgment across many jurisdictions require additional framing steps not covered here. The segmentation technique also assumes that primary records are publicly accessible; where material is restricted, the table will simply mark those rows as requiring alternative access routes.
The method does not replace the need to judge source authority. It only makes visible which claims have been checked against which records and which have not.
