From a question to an evidence trail

Start with a decision you cannot make yet

Imagine a school wants to replace part of its paved courtyard with planting beds. Some students expect less standing water after rain; others worry that the beds will become muddy. “Are gardens good?” cannot resolve this disagreement. The useful question concerns a particular outcome in a particular place: how might changing the surface affect where rainfall goes?

A research question connects an uncertainty with evidence you could collect. Before searching, write down the decision, the competing possibilities, and the observations that would distinguish them. A question such as “Which courtyard surfaces stay wet longest after similar rainfall?” can guide an initial investigation. It is narrower than asking whether the entire redesign will succeed, and the answer should be presented with that narrower scope.

Build a question with four boundaries

Specify the subject, location, comparison, and outcome. In our case these are surface type, the courtyard, paving versus planted soil, and visible water after rainfall. You might also need a time boundary: observations made ten minutes after rain answer a different question from observations made the next morning.

The final decision may require multiple smaller questions. Drainage observations cannot, on their own, establish cost, accessibility, or maintenance requirements.

Keep a source ledger, not a pile of links

A useful source record explains why you collected an item. Include its author or organization, title, publication date where available, exact location, main claim, evidence, and one limitation. A link alone makes it difficult to remember whether you saved the source for its measurements, a definition, or an example.

For the courtyard investigation, an Earth-science explanation can establish what infiltration means. A site plan can identify slopes and drains. Your observations can describe particular puddles. A contractor’s advertisement may describe a product but provide limited evidence about performance in your courtyard. These sources contribute different things; they should not be treated as interchangeable votes.

When a page summarizes another source, follow the reference. Five summaries of the same study are still dependent on one underlying study. Record the relationship so that repeated claims do not appear to be independent confirmation.

Separate observation, explanation, and recommendation

“A puddle remained beside the doorway at 9 a.m.” is an observation. “The drain may be obstructed” is a possible explanation. “Inspect the drain before changing the surface” is a recommendation. Each moves a step beyond the previous statement and needs a different kind of support.

Suppose the planted area dries sooner. That observation is consistent with greater infiltration, but the areas might also differ in slope, sun exposure, or the amount of rain reaching them. A careful report identifies those possibilities. It does not turn an uncontrolled comparison into proof that one surface caused the entire difference.

Try a three-column notebook page. Put what you saw in the first column, interpretations in the second, and proposed actions in the third. Add a question mark wherever the evidence leaves room for more than one explanation.

Write with attribution that can be checked

Read a short source passage, put it aside, and explain the meaning in your own sentence structure. Reopen the source to check that you preserved qualifications such as “may,” “under these conditions,” or “in this sample.” Changing a cautious statement into an absolute one is an inaccurate paraphrase even if every word is new.

Keep direct quotations visibly separate from notes. For a school report, follow the citation style your educator requests. At minimum, make it possible to locate the exact material: author or organization, title, date if available, and page or URL. Credit borrowed ideas as well as copied wording. Label your own calculations and observations so readers know where the evidence came from.

A worked evidence paragraph

During two observations after rainfall, water remained longer in the paved corner than in the planted strip. This suggests a difference in drainage under those conditions, but the areas also differ in slope and shade. An explanation of infiltration helps interpret the pattern; it does not establish the cause at this particular site. Before recommending a redesign, we would inspect the existing drain and compare more similar surfaces.

The paragraph states the observation first, narrows the inference, connects a concept, and names the next evidence needed. It makes progress without pretending the investigation is finished.

Now write your own paragraph using a different classroom question. Underline the observation, circle the inference, and identify the sentence that limits the conclusion. If those parts cannot be distinguished, revise before adding more sources.

Use the source desk with judgment

The interactive source desk on this site asks about authorship, evidence, relevance, and independent checking. It is a planning aid. Ticking every box cannot certify that a source is true, and a missing date does not automatically make every historical source useless. The value is in naming what you checked and what remains uncertain.

Return to the lesson library for focused practice on question design, source evaluation, and paraphrasing. For an extended task, produce a one-page evidence trail in which each major claim points to a source record or a clearly labeled observation.