Conduct a short research project from a self-generated question with multiple sources, credibility checks and standard citation

Learning goal: Conduct short research projects to answer a question, including a self-generated question, drawing on several sources and generating additional related, focused questions that allow multiple avenues of exploration. Gather relevant information from multiple print and digital sources using search terms effectively; assess the credibility and accuracy of each source; quote or paraphrase the data and conclusions of others while avoiding plagiarism and following a standard format for citation. Write routinely over extended time frames for research, reflection and revision, and over shorter time frames, for discipline-specific tasks, purposes and audiences.

A research question is generated, not assigned. The argument lesson ended with an open point: the garden could work if summer watering were solved. That gap is where a self-generated question comes from. "Is a garden a good idea?" is too broad to research in a short project: it names no criteria, so no source can answer it as asked, though narrowed to feasibility, cost or maintenance it becomes researchable. "How much water would eight raised beds at Ridgeview need between mid-June and mid-August, and who could supply it?" is focused: it names a quantity, a place, a period and a second question about people. A good short-project question can be answered in two to three weeks from several sources, and its answer would change what someone decides.

Generate additional related, focused questions. As sources are read, a research project branches. The watering question produces related questions that allow multiple avenues of exploration: Could rain barrels collect enough from the courtyard roofline? Which plants tolerate a two-week gap? What did Maple Hollow try in its second summer, and why did it fail? Is the courtyard water line shut off in summer, and who can request that it stay on? Each is narrower than the original and each could be a project of its own. Recording them as they arise is part of the research; choosing which to pursue is the researcher's judgment.

Gather information from multiple print and digital sources, using search terms effectively. A search engine returns what the words ask for. "garden" returns millions of pages; "raised bed watering schedule summer" returns pages about the question. Quotation marks hold a phrase together ("raised bed"); adding a place or a type of source narrows further (a school district's facilities pages, a university extension service). Print sources still matter: a library book on small-space vegetable gardening has an author, an editor and a date, and it will not change after you cite it. For this project, multiple means at least three, of more than one kind, because a single source cannot be checked against itself.

The four simulated sources used in the model. All four are invented for this lesson; none exists, and none may be cited outside it. Source 1, a print book chapter on watering raised beds, states that a 1.2-meter by 2.4-meter bed in hot weather needs 20 to 30 liters per watering, about three times a week. Source 2, the district's facilities web page, states that courtyard water lines are shut off from mid-June to mid-August unless a staff member requests otherwise. Source 3, a community gardening blog post with no author name and no date, states that "raised beds never need watering more than once a week." Source 4, a phone interview the student conducted with the school's custodian, who said the valve key hangs in the custodial closet and that no one has ever asked for the courtyard line to stay on in summer.

Assess the credibility and accuracy of each source. Credibility is about the source: who wrote it, what they know, when, and why. Accuracy is about the claim: does it agree with other sources and with what can be checked? Source 1 has a named author, a publisher and a date, and its figures are given as a range with a condition (hot weather), which is how careful sources speak. Source 2 is an official page of the organization that controls the water line, credible for that fact and for nothing about plants. Source 3 has no author, no date and a claim ("never") that contradicts Source 1; its purpose appears to be reassurance rather than information, and it should be set aside or cited only as an example of an unsupported claim. Source 4 is a primary source, the person with direct knowledge of the key and the line, credible for what he has seen and not for how much water beans need. Corroboration is the test: a claim that appears in two independent, credible sources is stronger than a claim in one.

Quote or paraphrase the data and conclusions of others, and avoid plagiarism. To quote is to reproduce the source's exact words inside quotation marks with a citation. To paraphrase is to restate the source's information in your own words and sentence structure, also with a citation. Plagiarism is presenting another's words, data or conclusions as your own, and it includes changing a few words of a sentence and dropping the citation. The test for a paraphrase is to close the source, write the idea from memory in your own structure, then reopen the source to check the facts. Data belong to the source that collected them: "eight beds would need about 600 liters a week" is the student's computation, but the 25 liters it rests on is Source 1's figure and must be cited, and the assumption that Ridgeview's beds are the size the guide describes is the student's and must be stated.

Follow a standard format for citation. Standard formats exist so that a reader can find the source. This lesson uses the MLA format, which many middle-school English classes use; if your class uses another, the same elements appear in a different order. Every entry records: who (the author, or the organization if there is no author; if neither, the entry begins with the title), what (the title, in italics for a whole work and in quotation marks for a part of one), where (the publisher, or the site that contains the page), when (the date of publication) and, for a digital source, its web address, plus the date you accessed it when the page gives no date of its own. In the text, a brief citation in parentheses (the author's last name, or the short title when there is no author, plus the page number if the source has pages) points to the full entry in the alphabetical list at the end. Missing elements are handled by rule, not by guessing: no author, start with the title; no date, omit it and give your access date; no page numbers, cite the name alone. Apply the format to every entry the same way.

Original model research paragraph, with its source list.

Eight raised beds at Ridgeview, if each is the 1.2-by-2.4-meter size the guide describes, would need roughly 600 liters of water a week during a hot spell, and the courtyard's water line is shut off for exactly the weeks when they would need it. A gardening guide gives 20 to 30 liters per watering for a bed 1.2 meters by 2.4 meters in hot weather, about three times a week (Whitfield 44); no second source giving a figure was found, so the range rests on that guide alone. Using the midpoint, 25 liters, eight beds watered three times a week require 8 × 25 × 3 = 600 liters (480 to 720 liters across the guide's range). The district's facilities page states that courtyard water lines "are shut off from mid-June to mid-August unless a staff member requests otherwise" (Ridgeview District, "Summer Facilities"), and the school's custodian confirmed in an interview that no such request has ever been made (Ansel). A community blog claims raised beds "never need watering more than once a week," but the post is unsigned, undated and contradicted by the guide, so it was not relied on ("Stop Overwatering"). The evidence suggests that the water supply is solvable by a request, but it says nothing about the labor of watering: with the line on, someone would still have to water eight beds three times a week, and no source names that person. That raises two further questions: whether a staff member would make the request, and who would do the watering.

Works Cited (MLA format; every source is invented for this lesson, so the web addresses are shown as placeholders). Ansel, Marcus. Telephone interview with the author. 12 May 2026. / Ridgeview District. "Summer Facilities." Ridgeview Local Schools, 3 Mar. 2026, [web address]. / "Stop Overwatering." Community Plot, [web address]. Accessed 10 May 2026. / Whitfield, Rina. Vegetables in Small Spaces. Greenrow Press, 2021.

Read the entries against the rule. The interview names the person, the kind of interview and its date. The district page has an organization as author, a page title in quotation marks, the site in italics, a date and an address. The blog post has no author, so it begins with its title, and no date, so the access date is given. The book has author, title, publisher and year; the page used appears in the in-text citation (Whitfield 44), not in the entry. The list is alphabetical by each entry's first element.

Write routinely over extended and shorter time frames. This project is an extended time frame: two to three weeks with time for research, reflection on what the sources do and do not show, and revision of the report. Inside it are shorter time frames: a research log entry written in a single sitting at the end of each work session (what you searched, what you found, what you now doubt), a one-day draft of the paragraph above, a same-day reflection after the interview. Discipline-specific means the writing follows the conventions of its field: a science-class report of the same project would give the liters in a table with units and a method section; a social-studies version would weigh who decides about the water line. Routine writing is how researchers keep track of their own thinking, and it is graded on being kept, not on being polished.

Worked narrowing of a question: "Should Ridgeview have a garden?" cannot be answered by sources as asked, because it names no criteria; a decision question becomes researchable once it says what would count. "How much water would eight beds need in summer?" can be: it names a quantity that sources give. Narrowing further, "How much water would eight beds need between mid-June and mid-August, and who could supply it?" adds the period the district page defines and a second, people-focused part that the interview answers. The related questions it generated (rain barrels, drought-tolerant plants, Maple Hollow's second summer, who can request the line stay on) were logged, and the last was pursued because the interview could answer it in one call.

Worked search-term sequence: "garden watering" returned general pages. "raised bed watering summer schedule" returned pages about the question. Adding "extension" found pages from university extension services, a source type known for named authors and dated, tested guidance; in the simulation, Source 1 was found in the library catalog by the search "raised bed vegetables" limited to books. The district page was found by searching the district site itself for "courtyard water." Each refinement traded quantity for relevance.

Worked credibility comparison: Source 1 (author, publisher, date, ranged figure with a stated condition) and Source 3 (no author, no date, an absolute claim with "never") disagree. Credibility favors Source 1 on every criterion. Accuracy is checked by corroboration: the student looked for a second guide that gave a figure; in the simulation none was found within the project's time, so the report rests its range on Source 1 alone and says so. Source 3 was cited only to show that it had been considered and set aside, which is honest reporting, not reliance.

Worked quote versus paraphrase versus plagiarism: Source 2's sentence: "Courtyard water lines are shut off from mid-June to mid-August unless a staff member requests otherwise." Quotation: reproduce it in quotation marks with the citation. Paraphrase: "The district turns off courtyard water for the summer unless a staff member asks that it stay on (Ridgeview District)." Plagiarism: "Courtyard water lines are turned off from mid-June until mid-August unless staff request otherwise," with no citation, changes only a few words and claims the sentence. The computation 8 × 25 × 3 = 600 is the student's own work, but 25 rests on Source 1 and is cited.

Research log (routine writing, short time frames): Over two weeks, write a log entry at the end of each research session, in a single sitting, of about 120 words: the question as it currently stands, the search terms tried and what each returned, each source found with its who, what, where and when, one sentence assessing its credibility, and any new related question the session generated. Entries are kept, not polished. At the end of the project, reread the log and write a 150-word reflection on how the question changed and why.

Independent performance (extended time frame): Conduct a short research project of two to three weeks on a self-generated question that arises from this unit or from a course you are taking (a science, social studies or health question is welcome, and the report should follow that discipline's conventions). Success criteria: a focused question that names what, where and when; at least three sources of at least two kinds, including at least one print source and at least one digital source; a search-term record; a credibility and accuracy assessment of every source, with one source set aside and the reason stated; at least three additional related, focused questions logged, one pursued; a 400-word report that quotes at least once and paraphrases at least twice, each with an in-text citation; no plagiarism, verified by the close-the-source paraphrase test; a works-cited list in MLA format (or the format your class uses), applied consistently; and the research log and reflection submitted with it. If your sources are real, cite them accurately and never invent a figure; if a figure cannot be found, write that it could not be found.