Research Understanding
Count the intersection before choosing the next reading set
Imagine a fictional collection of twelve records. Under supplied classification rules, eight match the topic and five count as current. Three meet both conditions. The exercise does not provide actual documents or define a real-world standard of currency. Its task is to recover the four groups implied by these counts and choose the group that satisfies a request for records that are both topical and current.
01
Recover the four groups without double-counting
The three records meeting both conditions are already included in the topic count of eight and the current count of five. Subtract three from eight to get five topic-only records. Subtract three from five to get two current-only records. The union contains five plus three plus two, or ten records. That leaves two of the twelve in neither group. As a check, the four disjoint groups total five plus three plus two plus two, which returns the collection size. Adding eight and five directly would count the shared three twice.
02
Match the selection rule to the actual question
A request for records that are both topical and current selects the three in the intersection. A request for either condition selects the ten in the union. These are different reading sets even though they start from the same collection. Make the connecting word explicit when recording the question. Neither selection establishes that the documents are correct, authoritative or enough to answer a substantive research question. The filters identify candidates according to supplied labels; a later reading stage must inspect the contents and their relevance to the claim being considered.
03
Preserve the rules alongside the candidate list
The reusable worksheet records the collection boundary, topic definition, currency rule, classification date, item identifiers and the requested combination of filters. Add the four group counts and retain any disputed labels for review. If a date threshold changes, recompute the groups rather than reusing the old total. If an item’s classification is unknown, record that state separately instead of quietly treating it as a failed condition. The example assumes every record has a supplied classification; a real collection may need that preliminary work before the arithmetic is meaningful.