Conduct short and sustained research: frame and adjust a question, search effectively, assess each source, synthesize, cite in a standard format and draw evidence from informational texts

Learning goal: Conduct short as well as more sustained research projects to answer a question, including a self-generated question, or solve a problem; narrow or broaden the inquiry when appropriate; synthesize multiple sources on the subject, demonstrating understanding of the subject under investigation; gather relevant information from multiple authoritative print and digital sources, using advanced searches effectively; assess the usefulness of each source in answering the research question; integrate information into the text selectively to maintain the flow of ideas, avoiding plagiarism and following a standard format for citation; and draw evidence from informational texts to support analysis, reflection and research.

Short and sustained research. A short research project answers a focused question in a lesson or two: "What is a dry dam?" answered from two authoritative sources and written up in a paragraph. A sustained project runs over weeks, answers a question that needs many sources and several stages, and usually changes shape as you learn. Both follow the same cycle: question, search, assess, read and record, synthesize, write, cite. Research can answer a question (how does something work, what happened, why) or solve a problem (which design, which policy, which method best meets a need).

Research to solve a problem. Problem-solving research is organized around a decision rather than an explanation. Define the problem and the criteria a solution must meet: what it must achieve, what it may cost, and what rules or conditions limit it. List the candidate solutions. Gather evidence on how each candidate performs against each criterion, from sources that can speak to it. Then synthesize by criterion, comparing the candidates side by side, and recommend the one that best meets the criteria, stating what the evidence cannot settle. The same cycle of question, search, assess, synthesize and cite applies; what changes is that every source is read for what it shows about an option against a criterion. A problem-solving question can also be too broad or too narrow: "How can floods be prevented?" names no place, no options and no criteria, while "Which flood-protection method would best protect one river neighborhood within its town's budget?" can be researched.

The question, and when to narrow or broaden it. A good research question is open (it cannot be answered yes or no without explanation), researchable (sources exist that can answer it) and sized to the project. Self-generated questions are often the best, because curiosity keeps a sustained project alive; they also tend to start too broad. The history lessons' story of a fictional town flooded in 1913 might prompt the self-generated question "How can floods be prevented?" That is too broad: it covers every river, method and century, and no report could answer it. Narrow it by method and scope: "How does a dry dam reduce flooding downstream?" Now a short project can answer it. Narrowing too far also fails: "What was the width of the outlet conduits at one particular dam?" is a single fact, not an inquiry, and may be unfindable. Broaden it back to a question that needs explanation. Narrow or broaden as the evidence demands: if your sources show that dry dams work only as part of a system with levees and channel improvements, the question may need to broaden to "How do dry dams work together with levees to reduce flooding?"

Gathering from authoritative print and digital sources. Authoritative sources are those whose authors or publishers have the expertise and accountability to be trusted on the topic: government agencies responsible for flood control, engineering textbooks, university extension services, historical societies and archives, established encyclopedias, peer-reviewed journals. Use multiple sources and more than one kind, print and digital, so that no single source's errors or slant go unchecked.

Using advanced searches effectively. A plain keyword search ranks results by the engine's own relevance signals, which often favor popular pages over the ones most useful for your question. Advanced search techniques narrow the results to what answers the question. Quotation marks around a phrase, "dry dam", ask for that exact phrase rather than pages that mention "dry" and "dam" separately (engines usually, though not always strictly, honor this). The site: operator limits results to one domain or kind of domain: site:gov returns pages on .gov addresses, which in the United States are used by many government bodies, and site:edu returns pages from .edu addresses, used mainly by United States colleges and universities. Know what a domain filter leaves out: the United States Army Corps of Engineers, the country's main federal flood-control builder, publishes on an army.mil address, and Ohio's Miami Conservancy District publishes on a .org address, so site:gov would miss both. A minus sign before a word excludes pages that contain that word, not pages on a subject: "dry dam" -beaver drops results containing "beaver," which cuts most beaver-dam pages but would also drop a useful engineering page that happens to mention beavers. Many engines also offer a filetype: operator (filetype:pdf returns PDF files, the format of many reports and technical documents) and date-range filters. Most major search engines support quotation marks and site:; support for other operators varies, so check the search engine's help page. Library databases add filters for subject, date, publication type and peer review, and often give better results than the open web for sustained projects.

Assessing the usefulness of each source. Usefulness is judged against your question, not in general. Ask of each source: Relevance: does it address the question or only the general topic? Authority: who wrote it, with what expertise, and who published it? Accuracy: does it give evidence, and can its claims be checked against other sources? Currency: is it recent enough for the question? For history, an older source may be essential; for current engineering practice, it may be outdated. Purpose: does it aim to inform, to persuade or to sell? Level: is it written at a depth you can use? A source can be authoritative and still not useful for your question: an engineering handbook on dam spillway design is reliable, but if your question is how dry dams reduce downstream flooding, most of it is not relevant.

Synthesizing multiple sources. Synthesis is not a summary of source one followed by a summary of source two. It is organized by idea: for each point your answer needs, you bring in the sources that speak to it (sometimes one, sometimes several), noting where they agree, where one adds what another lacks, and where they conflict. Synthesis demonstrates understanding of the subject under investigation, because only a writer who understands the subject can see how separate sources fit together. A synthesis matrix, with ideas as rows and sources as columns, shows at a glance where each idea is supported.

Integrating information selectively and maintaining the flow of ideas. Use the smallest piece of a source that does the job. Quote when the exact words matter, such as a definition or a historical voice. Paraphrase when the idea matters more than the wording. Introduce borrowed ideas with a signal phrase that names the source whenever the reader needs to see whose idea it is, make every attribution unambiguous, then connect the idea to your own point, so that your reasoning, not the sources, leads each paragraph.

Avoiding plagiarism. Plagiarism is presenting another person's words or ideas as your own. Quoted words need quotation marks and a citation. Paraphrased ideas need a citation even though the words are yours. Patchwriting, copying a source's sentence and swapping a few words, is not an acceptable paraphrase even when a citation is given, because the sentence structure and most of the words are still the source's: the citation credits the idea, but the borrowed wording, without quotation marks, is presented as yours. Whether a particular case counts as plagiarism is decided by your school's academic-honesty policy, so read it; the remedy is the same either way. Either quote the sentence exactly, with quotation marks and a citation, or read, close the source, write the idea from understanding, then check the original for accuracy and cite it. Common knowledge that appears in many general sources, such as the fact that water flows downhill, needs no citation.

Following a standard format for citation. Disciplines use standard formats so that any reader can find the source. MLA style (9th edition) is common in English and other humanities classes; APA style (7th edition) in psychology, education and many social sciences; historians most often use Chicago style (notes and bibliography); and the natural sciences mostly use their own discipline styles. Here are three formats applied to the invented Ashby essay from the history lessons. The source is fictional; the entries show format only. Titles shown in italics are italicized in the finished list.

MLA, in the list titled Works Cited: Ashby, R. M. "Water and Blame in a River Town." County Historical Journal, vol. 12, 1988, pp. 4–19. In-text: (Ashby 7). Article title in quotation marks and title case; add the issue number (no. 2) after the volume when the journal has one.

APA, in the list titled References: Ashby, R. M. (1988). Water and blame in a river town. County Historical Journal, 12, 4–19. In-text: (Ashby, 1988, p. 7). Article title in sentence case with no quotation marks; journal title and volume in italics; add the issue number in parentheses after the volume, and a DOI at the end, when the source has them.

Chicago (notes and bibliography), bibliography entry: Ashby, R. M. "Water and Blame in a River Town." County Historical Journal 12 (1988): 4–19. First footnote citing page 7: R. M. Ashby, "Water and Blame in a River Town," County Historical Journal 12 (1988): 7.

Follow whichever format your teacher names, consistently, and record full citation details while you research so you never have to reconstruct them.

Drawing evidence from informational texts to support analysis, reflection and research. Evidence for analysis supports a claim about how something works or what something means. Evidence for reflection supports a considered judgment about your own thinking: "I first assumed dams hold water permanently; the definition in the agency source showed me a dry dam is normally empty." Evidence for research answers the research question. In all three, draw specific evidence (the precise sentence, figure or definition), name its source, and explain how it supports your point.

A fully worked model. Question: How does a dry dam reduce flooding downstream? The synthesis below draws on three short excerpts written for this lesson to stand for three kinds of source (they are invented, not quotations from real publications), plus one real page listed in this lesson's references. Source A (invented agency page): "A dry dam has no permanent pool. In ordinary weather the river flows freely through conduits at the base of the dam." Source C (invented engineering textbook): "Outlet conduits limit discharge so that floodwater is temporarily detained upstream of the structure." Source D (invented historical-society article): "Between floods, the land in the storage basins was leased to farmers." Real source: the Miami Conservancy District's page on its flood-protection system says the storage land upstream of its dams is used mainly for parkland and farmland. Synthesis paragraph: "A dry dam, unlike a conventional reservoir dam, normally holds no lake; in ordinary weather the river passes through conduits at its base (Source A). During a flood, those conduits let only a limited flow through, so the excess water is held temporarily behind the dam (Source C) and must pass downstream later. Because the storage land is dry most of the time, it can be farmed between floods, as one historical account and the Miami Conservancy District both report (Source D; Miami Conservancy District). In a simplified picture that ignores small losses to evaporation and seepage, the dam therefore does not remove the flood's water; it lowers the peak flow downstream by spreading the same water over a longer time." The paragraph is organized by idea, not by source, and names the source of each borrowed point: the first two sentences each take one step of the explanation from the source that supplies it (A, then C), the third combines two sources on a single point (Source D and the Miami Conservancy District), and the last is the writer's own inference connecting them. The synthesis happens across the paragraph; not every sentence needs more than one source.

Worked problem-solving research plan: Problem (for the fictional town of the history lessons): "Which flood-protection method would best protect the Lower Ward, given what the village can afford?" Criteria: how much flood protection it gives, what it costs, and what effects it has upstream or downstream. Candidates: a levee along the ward, a wider and cleared river channel, or an upstream dry dam. Sources: an engineering text on each method's effect on peak flow, an agency page on costs and upkeep, and a historical account of a real system that combined methods, such as the Miami Conservancy District's dams, levees and channel improvements. Synthesis matrix: candidates as rows, criteria as columns, a source in every filled cell. Recommendation: the candidate that meets the criteria best, with the cells left empty named as limits. The plan differs from an explanatory project because it ends in a decision and every source is read against a criterion.

Worked narrowing and broadening: Start: "What causes floods?" Too broad: every climate, river and cause. Narrowed: "How does a dry dam reduce flooding downstream?" Researchable in a short project. Over-narrowed while searching: "How tall is the dam at one particular site?" A single fact that one record can answer and that needs no explanation. (Compare "When was the first dry dam built?": that could become a real inquiry, because it requires defining a dry dam and weighing competing claims, but it is a different project from the one that asks how dry dams work.) Broadened back when the sources show that dams are paired with levees: "How do dry dams and levees work together to reduce flood damage?" This fits a sustained project, because it needs sources on both methods and on how they interact.

Worked advanced search: A plain search, dry dam flooding, may return pages about beaver dams, droughts and news stories. Revised search: "dry dam" flood -beaver site:gov. The quotation marks ask for the exact phrase, -beaver excludes pages containing the word "beaver" (most beaver-dam pages, and any useful page that mentions beavers), and site:gov limits the results to .gov addresses, where many government agencies publish. Because that filter also removes authoritative flood-control sources on other domains (army.mil, .org), a careful researcher runs a second search with site:mil or without the domain filter and compares. A second search for technical detail adds filetype:pdf, because many technical reports are published as PDF files. If an operator does not work in your search engine, check its help page or use the engine's advanced-search form.

Worked source assessment and paraphrase check: Candidate sources for the dry-dam question, described by type with the details a researcher would record (all invented for this lesson): (A) a government flood-control agency page explaining how different dams work, updated last year, with diagrams and an engineering contact; (B) a real-estate company blog post titled "Is Your New Home Flood-Proof?", with no author named, no sources cited and a link to listings; (C) an engineering textbook chapter on detention structures by two university engineers; (D) a historical society article on a regional flood-control project, citing archival records. Assessment, given those details: A is authoritative, relevant and recently updated, so highly useful, though its claims should still be checked against another source. B has a sales purpose and, as described, no named author or cited evidence; useful only for leads to check elsewhere. A different blog post with a named expert and cited evidence could earn a higher rating, which is why the judgment rests on the details, not the type. C is authoritative and precise, useful for mechanism, though some of it exceeds the question. D is authoritative for history and useful if the question includes how a system was built. Paraphrase check: source sentence (invented): "Outlet conduits limit discharge so that floodwater is temporarily detained upstream of the structure." Patchwritten: "Outlet pipes limit discharge so floodwater is temporarily held upstream of the structure." Too close to count as a paraphrase, even with a citation: the structure and most of the words are still the source's, with no quotation marks. Either quote the sentence exactly and cite it, or recast it. Proper paraphrase: "Because the openings at the base let only so much water through, the rest waits behind the dam until the flood passes" plus a citation.

Synthesis matrix (representation): For your research question, build a matrix with the main ideas your answer needs as rows (for example: what a dry dam is; how it controls flow; its limits; its history in a region) and at least four sources as columns. In each cell, write a short paraphrase with a page or section reference, or leave it empty. Add a final row assessing each source's usefulness in one line (relevance, authority, accuracy, currency, purpose), and a column with your own synthesis sentence for each idea.

Independent performance: Complete a short research project on a self-generated question from your history or science unit. Success criteria: (1) a research question and a record of how you narrowed or broadened it, with the reason; (2) at least one advanced search string you used, explained; (3) at least four sources, print and digital, each with a one-sentence usefulness assessment; (4) a 400- to 550-word report organized by idea, not by source, with at least two sentences of genuine synthesis; (5) quotations and paraphrases integrated selectively with signal phrases, and no patchwriting; (6) a works-cited or reference list in one standard format, with matching in-text citations; (7) a closing reflection of 60 to 100 words that draws on specific evidence to explain how your understanding changed.

Sustained research project (over three to four weeks): Extend the question, or broaden it as the worked example does ("How do dry dams and levees work together to reduce flood damage?"), into a sustained project with dated checkpoints. Week 1: a question proposal and a search log of at least five search strings, noting which returned useful sources and how you changed the question. Week 2: a synthesis matrix with at least six sources, print and digital, each with a usefulness note, and a conference with your teacher or a peer about whether to narrow or broaden again. Week 3: a full draft of 1,000 to 1,500 words organized by idea, with citations in one standard format. Week 4: a revision that answers the feedback, a complete works-cited or reference list, and a 150-word research reflection naming one source that changed your thinking and one question you would pursue next. Success criteria: every checkpoint submitted on its date; at least one recorded change to the question with its reason; synthesis in every body section; no patchwriting; citations that match the list.